diff --git a/feed_weekly.xml b/feed_weekly.xml index dcd6c43..c3e6eb4 100644 --- a/feed_weekly.xml +++ b/feed_weekly.xml @@ -7,741 +7,741 @@ http://www.rssboard.org/rss-specification python-feedgen en - Mon, 15 Dec 2025 00:13:10 +0000 + Mon, 22 Dec 2025 00:13:11 +0000 - ThreadWeaver: Adaptive Threading for Efficient Parallel Reasoning in Language Models - https://arxiv.org/abs/2512.07843 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.07843.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Long Lian, Sida Wang, Felix Juefei-Xu, Tsu-Jui Fu, Xiuyu Li, Adam Yala, Trevor Darrell, Alane Suhr, Yuandong Tian, Xi Victoria Lin</p><p><b>Upvotes:</b> 19</p><p><b>Summary:</b> Scaling inference-time computation has enabled Large Language Models (LLMs) to achieve strong reasoning performance, but inherently sequential decoding leads to substantial latency, especially on complex tasks. Recent work on adaptive parallel reasoning aims to improve inference efficiency by decomposing the problem-solving process into concurrent reasoning threads when beneficial. However, existing methods on realistic tasks are either limited to supervised behavior cloning or exhibit significant accuracy drops compared to widely-used sequential long chain-of-thought (CoT) baselines. Moreover, many require customized inference engines, complicating deployment. We introduce ThreadWeaver, a framework for adaptive parallel reasoning that achieves accuracy on par with popular sequential reasoning models of comparable size while significantly reducing inference latency. ThreadWeaver's performance stems from three key innovations: 1) a two-stage parallel trajectory generator that produces large-scale, high-quality CoT data with parallel annotations for supervised fine-tuning; 2) a trie-based training-inference co-design that enables parallel reasoning on any off-the-shelf autoregressive inference engine without modifying position embeddings or KV caches; and 3) a parallelization-aware reinforcement learning framework that teaches the model to balance accuracy with effective parallelization. Across six challenging mathematical reasoning benchmarks, ThreadWeaver trained atop Qwen3-8B achieves accuracy comparable to cutting-edge sequential reasoning models (71.9% on average and 79.9% on AIME24) while delivering up to 1.53x average speedup in token latency, establishing a new Pareto frontier between accuracy and efficiency.</p> - https://arxiv.org/abs/2512.07843 - Mon, 24 Nov 2025 18:55:59 +0000 + MeshSplatting: Differentiable Rendering with Opaque Meshes + https://arxiv.org/abs/2512.06818 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.06818.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jan Held, Sanghyun Son, Renaud Vandeghen, Daniel Rebain, Matheus Gadelha, Yi Zhou, Anthony Cioppa, Ming C. Lin, Marc Van Droogenbroeck, Andrea Tagliasacchi</p><p><b>Upvotes:</b> 10</p><p><b>Summary:</b> Primitive-based splatting methods like 3D Gaussian Splatting have revolutionized novel view synthesis with real-time rendering. However, their point-based representations remain incompatible with mesh-based pipelines that power AR/VR and game engines. We present MeshSplatting, a mesh-based reconstruction approach that jointly optimizes geometry and appearance through differentiable rendering. By enforcing connectivity via restricted Delaunay triangulation and refining surface consistency, MeshSplatting creates end-to-end smooth, visually high-quality meshes that render efficiently in real-time 3D engines. On Mip-NeRF360, it boosts PSNR by +0.69 dB over the current state-of-the-art MiLo for mesh-based novel view synthesis, while training 2x faster and using 2x less memory, bridging neural rendering and interactive 3D graphics for seamless real-time scene interaction. The project page is available at https://meshsplatting.github.io/.</p> + https://arxiv.org/abs/2512.06818 + Sun, 07 Dec 2025 12:31:04 +0000 - VQRAE: Representation Quantization Autoencoders for Multimodal Understanding, Generation and Reconstruction - https://arxiv.org/abs/2511.23386 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2511.23386.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Sinan Du, Jiahao Guo, Bo Li, Shuhao Cui, Zhengzhuo Xu, Yifu Luo, Yongxian Wei, Kun Gai, Xinggang Wang, Kai Wu, Chun Yuan</p><p><b>Upvotes:</b> 12</p><p><b>Summary:</b> Unifying multimodal understanding, generation and reconstruction representation in a single tokenizer remains a key challenge in building unified models. Previous research predominantly attempts to address this in a dual encoder paradigm, e.g., utilizing the separate encoders for understanding and generation respectively or balancing semantic representations and low-level features with contrastive loss. In this paper, we propose VQRAE, a Vector Quantization version of Representation AutoEncoders, which pioneers the first exploration in unified representation to produce Continuous semantic features for image understanding and Discrete tokens for visual generation within a unified tokenizer. Specifically, we build upon pretrained vision foundation models with a symmetric ViT decoder and adopt a two-stage training strategy: first, it freezes the encoder and learns a high-dimensional semantic VQ codebook with pixel reconstruction objective; then jointly optimizes the encoder with self-distillation constraints. This design enables negligible semantic information for maintaining the ability of multimodal understanding, discrete tokens that are compatible for generation and fine-grained reconstruction. Besides, we identify the intriguing property in quantizing semantic encoders that rely on high-dimensional codebook in contrast to the previous common practice of low-dimensional codebook in image reconstruction. The semantic VQ codebook can achieve a 100% utilization ratio at a dimension of 1536. VQRAE presents competitive performance on several benchmarks of visual understanding, generation and reconstruction with promising scaling property in the autoregressive paradigm for its discrete merits.</p> - https://arxiv.org/abs/2511.23386 - Fri, 28 Nov 2025 17:26:34 +0000 + Aesthetic Alignment Risks Assimilation: How Image Generation and Reward Models Reinforce Beauty Bias and Ideological "Censorship" + https://arxiv.org/abs/2512.11883 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.11883.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Wenqi Marshall Guo, Qingyun Qian, Khalad Hasan, Shan Du</p><p><b>Upvotes:</b> 6</p><p><b>Summary:</b> Over-aligning image generation models to a generalized aesthetic preference conflicts with user intent, particularly when ``anti-aesthetic" outputs are requested for artistic or critical purposes. This adherence prioritizes developer-centered values, compromising user autonomy and aesthetic pluralism. We test this bias by constructing a wide-spectrum aesthetics dataset and evaluating state-of-the-art generation and reward models. We find that aesthetic-aligned generation models frequently default to conventionally beautiful outputs, failing to respect instructions for low-quality or negative imagery. Crucially, reward models penalize anti-aesthetic images even when they perfectly match the explicit user prompt. We confirm this systemic bias through image-to-image editing and evaluation against real abstract artworks.</p> + https://arxiv.org/abs/2512.11883 + Tue, 09 Dec 2025 00:24:29 +0000 - Visual Generation Tuning - https://arxiv.org/abs/2511.23469 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2511.23469.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jiahao Guo, Sinan Du, Jingfeng Yao, Wenyu Liu, Bo Li, Haoxiang Cao, Kun Gai, Chun Yuan, Kai Wu, Xinggang Wang</p><p><b>Upvotes:</b> 13</p><p><b>Summary:</b> Large Vision Language Models (VLMs) effectively bridge the modality gap through extensive pretraining, acquiring sophisticated visual representations aligned with language. However, it remains underexplored whether these representations, optimized for multimodal understanding tasks, harbor an inherent potential for visual generation. In this paper, we propose VGT, Visual Generation Tuning, a novel paradigm designed to stimulate the underlying capabilities of visual generation within any vision language models. By performing efficient visual generation tuning on well-pretrained VLMs, we significantly mitigate the alignment costs and accelerate the convergence of autoregressive modeling in the continuous space (20x speedup). Specifically, we dismiss the entangled pixel-level VAEs designed for diffusion transformers and formulate VGT-AE through aligning the semantic encoders from pretrained VLMs with the latent representations of pixel decoders. In image reconstruction tasks, we achieve 26.67 PSNR and 0.50 rFID at a 28x compression ratio, outperforming specialized VAEs; in visual generation tasks, we achieve state-of-the-art outcomes among autoregressive models, 0.77 on GenEval and 78.73 on DPG-Bench. Furthermore, our proposed VGT showcases significant scaling promise and is versatile for endowing any VLMs trained for multimodal understanding with the capabilities of visual generation, which paves the new avenue to explore next-generation unified multimodal foundation models. Models and codes are available at https://github.com/hustvl/VGT.</p> - https://arxiv.org/abs/2511.23469 - Fri, 28 Nov 2025 18:57:13 +0000 + EgoX: Egocentric Video Generation from a Single Exocentric Video + https://arxiv.org/abs/2512.08269 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.08269.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Taewoong Kang, Kinam Kim, Dohyeon Kim, Minho Park, Junha Hyung, Jaegul Choo</p><p><b>Upvotes:</b> 101</p><p><b>Summary:</b> Egocentric perception enables humans to experience and understand the world directly from their own point of view. Translating exocentric (third-person) videos into egocentric (first-person) videos opens up new possibilities for immersive understanding but remains highly challenging due to extreme camera pose variations and minimal view overlap. This task requires faithfully preserving visible content while synthesizing unseen regions in a geometrically consistent manner. To achieve this, we present EgoX, a novel framework for generating egocentric videos from a single exocentric input. EgoX leverages the pretrained spatio temporal knowledge of large-scale video diffusion models through lightweight LoRA adaptation and introduces a unified conditioning strategy that combines exocentric and egocentric priors via width and channel wise concatenation. Additionally, a geometry-guided self-attention mechanism selectively attends to spatially relevant regions, ensuring geometric coherence and high visual fidelity. Our approach achieves coherent and realistic egocentric video generation while demonstrating strong scalability and robustness across unseen and in-the-wild videos.</p> + https://arxiv.org/abs/2512.08269 + Tue, 09 Dec 2025 05:53:39 +0000 - RealGen: Photorealistic Text-to-Image Generation via Detector-Guided Rewards - https://arxiv.org/abs/2512.00473 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.00473.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Junyan Ye, Leiqi Zhu, Yuncheng Guo, Dongzhi Jiang, Zilong Huang, Yifan Zhang, Zhiyuan Yan, Haohuan Fu, Conghui He, Weijia Li</p><p><b>Upvotes:</b> 24</p><p><b>Summary:</b> With the continuous advancement of image generation technology, advanced models such as GPT-Image-1 and Qwen-Image have achieved remarkable text-to-image consistency and world knowledge However, these models still fall short in photorealistic image generation. Even on simple T2I tasks, they tend to produce " fake" images with distinct AI artifacts, often characterized by "overly smooth skin" and "oily facial sheens". To recapture the original goal of "indistinguishable-from-reality" generation, we propose RealGen, a photorealistic text-to-image framework. RealGen integrates an LLM component for prompt optimization and a diffusion model for realistic image generation. Inspired by adversarial generation, RealGen introduces a "Detector Reward" mechanism, which quantifies artifacts and assesses realism using both semantic-level and feature-level synthetic image detectors. We leverage this reward signal with the GRPO algorithm to optimize the entire generation pipeline, significantly enhancing image realism and detail. Furthermore, we propose RealBench, an automated evaluation benchmark employing Detector-Scoring and Arena-Scoring. It enables human-free photorealism assessment, yielding results that are more accurate and aligned with real user experience. Experiments demonstrate that RealGen significantly outperforms general models like GPT-Image-1 and Qwen-Image, as well as specialized photorealistic models like FLUX-Krea, in terms of realism, detail, and aesthetics. The code is available at https://github.com/yejy53/RealGen.</p> - https://arxiv.org/abs/2512.00473 - Sat, 29 Nov 2025 12:52:26 +0000 + VLSA: Vision-Language-Action Models with Plug-and-Play Safety Constraint Layer + https://arxiv.org/abs/2512.11891 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.11891.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Songqiao Hu, Zeyi Liu, Shuang Liu, Jun Cen, Zihan Meng, Xiao He</p><p><b>Upvotes:</b> 8</p><p><b>Summary:</b> Vision-Language-Action (VLA) models have demonstrated remarkable capabilities in generalizing across diverse robotic manipulation tasks. However, deploying these models in unstructured environments remains challenging due to the critical need for simultaneous task compliance and safety assurance, particularly in preventing potential collisions during physical interactions. In this work, we introduce a Vision-Language-Safe Action (VLSA) architecture, named AEGIS, which contains a plug-and-play safety constraint (SC) layer formulated via control barrier functions. AEGIS integrates directly with existing VLA models to improve safety with theoretical guarantees, while maintaining their original instruction-following performance. To evaluate the efficacy of our architecture, we construct a comprehensive safety-critical benchmark SafeLIBERO, spanning distinct manipulation scenarios characterized by varying degrees of spatial complexity and obstacle intervention. Extensive experiments demonstrate the superiority of our method over state-of-the-art baselines. Notably, AEGIS achieves a 59.16% improvement in obstacle avoidance rate while substantially increasing the task execution success rate by 17.25%. To facilitate reproducibility and future research, we make our code, models, and the benchmark datasets publicly available at https://vlsa-aegis.github.io/.</p> + https://arxiv.org/abs/2512.11891 + Tue, 09 Dec 2025 16:53:44 +0000 - See, Hear, and Understand: Benchmarking Audiovisual Human Speech Understanding in Multimodal Large Language Models - https://arxiv.org/abs/2512.02231 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.02231.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Le Thien Phuc Nguyen, Zhuoran Yu, Samuel Low Yu Hang, Subin An, Jeongik Lee, Yohan Ban, SeungEun Chung, Thanh-Huy Nguyen, JuWan Maeng, Soochahn Lee, Yong Jae Lee</p><p><b>Upvotes:</b> 7</p><p><b>Summary:</b> Multimodal large language models (MLLMs) are expected to jointly interpret vision, audio, and language, yet existing video benchmarks rarely assess fine-grained reasoning about human speech. Many tasks remain visually solvable or only coarsely evaluate speech, offering limited insight into whether models can align who speaks, what is said, and when it occurs. We introduce AV-SpeakerBench, a curated benchmark of 3,212 multiple-choice questions focused on speaker-centric audiovisual reasoning in real-world videos. It features: (1) a speaker-centered formulation that treats speakers-not scenes-as the core reasoning unit; (2) fusion-grounded question design embedding audiovisual dependencies into question semantics; and (3) expert-curated annotations ensuring temporal precision and cross-modal validity. Comprehensive evaluations show that the Gemini family consistently outperforms open-source systems, with Gemini 2.5 Pro achieving the best results. Among open models, Qwen3-Omni-30B approaches Gemini 2.0 Flash but remains far behind Gemini 2.5 Pro, primarily due to weaker audiovisual fusion rather than visual perception. We believe AV-SpeakerBench establishes a rigorous foundation for advancing fine-grained audiovisual reasoning in future multimodal systems.</p> - https://arxiv.org/abs/2512.02231 - Mon, 01 Dec 2025 21:57:26 +0000 + VABench: A Comprehensive Benchmark for Audio-Video Generation + https://arxiv.org/abs/2512.09299 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.09299.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Daili Hua, Xizhi Wang, Bohan Zeng, Xinyi Huang, Hao Liang, Junbo Niu, Xinlong Chen, Quanqing Xu, Wentao Zhang</p><p><b>Upvotes:</b> 7</p><p><b>Summary:</b> Recent advances in video generation have been remarkable, enabling models to produce visually compelling videos with synchronized audio. While existing video generation benchmarks provide comprehensive metrics for visual quality, they lack convincing evaluations for audio-video generation, especially for models aiming to generate synchronized audio-video outputs. To address this gap, we introduce VABench, a comprehensive and multi-dimensional benchmark framework designed to systematically evaluate the capabilities of synchronous audio-video generation. VABench encompasses three primary task types: text-to-audio-video (T2AV), image-to-audio-video (I2AV), and stereo audio-video generation. It further establishes two major evaluation modules covering 15 dimensions. These dimensions specifically assess pairwise similarities (text-video, text-audio, video-audio), audio-video synchronization, lip-speech consistency, and carefully curated audio and video question-answering (QA) pairs, among others. Furthermore, VABench covers seven major content categories: animals, human sounds, music, environmental sounds, synchronous physical sounds, complex scenes, and virtual worlds. We provide a systematic analysis and visualization of the evaluation results, aiming to establish a new standard for assessing video generation models with synchronous audio capabilities and to promote the comprehensive advancement of the field.</p> + https://arxiv.org/abs/2512.09299 + Wed, 10 Dec 2025 03:57:29 +0000 - From Imitation to Discrimination: Toward A Generalized Curriculum Advantage Mechanism Enhancing Cross-Domain Reasoning Tasks - https://arxiv.org/abs/2512.02580 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.02580.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Changpeng Yang, Jinyang Wu, Yuchen Liu, Shuai Zhang, Yang Li, Qiliang Liang, Hongzhen Wang, Shuai Nie, Jiaming Xu, Runyu Shi, Ying Huang, Guoquan Zhang</p><p><b>Upvotes:</b> 27</p><p><b>Summary:</b> Reinforcement learning has emerged as a paradigm for post-training large language models, boosting their reasoning capabilities. Such approaches compute an advantage value for each sample, reflecting better or worse performance than expected, thereby yielding both positive and negative signals for training. However, the indiscriminate mixing of the two signals in existing methods, especially from the early stages, may lead to ambiguous guidance and limited gains. To address this issue, we propose **CAPO** (**C**urriculum **A**dvantage **P**olicy **O**ptimization), an adaptive curriculum mechanism based on advantage signals. The proposed mechanism bootstraps imitation learning with positive-only advantage samples to establish robust foundations, and subsequently introduces negative signals to cultivate discriminative capabilities, thereby improving generalization across complex scenarios. Compatible with diverse optimization methods including GRPO, PPO, RLOO, and Reinforce++, our method consistently achieves stable and significant improvements in mathematical reasoning tasks, and further generalizes effectively to multimodal Graphical User Interface (GUI) reasoning scenarios, establishing itself as a versatile and robust optimization framework.</p> - https://arxiv.org/abs/2512.02580 - Tue, 02 Dec 2025 09:48:57 +0000 + LLaDA2.0: Scaling Up Diffusion Language Models to 100B + https://arxiv.org/abs/2512.15745 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.15745.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Tiwei Bie, Maosong Cao, Kun Chen, Lun Du, Mingliang Gong, Zhuochen Gong, Yanmei Gu, Jiaqi Hu, Zenan Huang, Zhenzhong Lan, Chengxi Li, Chongxuan Li, Jianguo Li, Zehuan Li, Huabin Liu, Ling Liu, Guoshan Lu, Xiaocheng Lu, Yuxin Ma, Jianfeng Tan, Lanning Wei, Ji-Rong Wen, Yipeng Xing, Xiaolu Zhang, Junbo Zhao, Da Zheng, Jun Zhou, Junlin Zhou, Zhanchao Zhou, Liwang Zhu, Yihong Zhuang</p><p><b>Upvotes:</b> 58</p><p><b>Summary:</b> This paper presents LLaDA2.0 -- a tuple of discrete diffusion large language models (dLLM) scaling up to 100B total parameters through systematic conversion from auto-regressive (AR) models -- establishing a new paradigm for frontier-scale deployment. Instead of costly training from scratch, LLaDA2.0 upholds knowledge inheritance, progressive adaption and efficiency-aware design principle, and seamless converts a pre-trained AR model into dLLM with a novel 3-phase block-level WSD based training scheme: progressive increasing block-size in block diffusion (warm-up), large-scale full-sequence diffusion (stable) and reverting back to compact-size block diffusion (decay). Along with post-training alignment with SFT and DPO, we obtain LLaDA2.0-mini (16B) and LLaDA2.0-flash (100B), two instruction-tuned Mixture-of-Experts (MoE) variants optimized for practical deployment. By preserving the advantages of parallel decoding, these models deliver superior performance and efficiency at the frontier scale. Both models were open-sourced.</p> + https://arxiv.org/abs/2512.15745 + Wed, 10 Dec 2025 09:26:18 +0000 - PaCo-RL: Advancing Reinforcement Learning for Consistent Image Generation with Pairwise Reward Modeling - https://arxiv.org/abs/2512.04784 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.04784.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Bowen Ping, Chengyou Jia, Minnan Luo, Changliang Xia, Xin Shen, Zhuohang Dang, Hangwei Qian</p><p><b>Upvotes:</b> 24</p><p><b>Summary:</b> Consistent image generation requires faithfully preserving identities, styles, and logical coherence across multiple images, which is essential for applications such as storytelling and character design. Supervised training approaches struggle with this task due to the lack of large-scale datasets capturing visual consistency and the complexity of modeling human perceptual preferences. In this paper, we argue that reinforcement learning (RL) offers a promising alternative by enabling models to learn complex and subjective visual criteria in a data-free manner. To achieve this, we introduce PaCo-RL, a comprehensive framework that combines a specialized consistency reward model with an efficient RL algorithm. The first component, PaCo-Reward, is a pairwise consistency evaluator trained on a large-scale dataset constructed via automated sub-figure pairing. It evaluates consistency through a generative, autoregressive scoring mechanism enhanced by task-aware instructions and CoT reasons. The second component, PaCo-GRPO, leverages a novel resolution-decoupled optimization strategy to substantially reduce RL cost, alongside a log-tamed multi-reward aggregation mechanism that ensures balanced and stable reward optimization. Extensive experiments across the two representative subtasks show that PaCo-Reward significantly improves alignment with human perceptions of visual consistency, and PaCo-GRPO achieves state-of-the-art consistency performance with improved training efficiency and stability. Together, these results highlight the promise of PaCo-RL as a practical and scalable solution for consistent image generation. The project page is available at https://x-gengroup.github.io/HomePage_PaCo-RL/.</p> - https://arxiv.org/abs/2512.04784 - Tue, 02 Dec 2025 13:39:03 +0000 + MentraSuite: Post-Training Large Language Models for Mental Health Reasoning and Assessment + https://arxiv.org/abs/2512.09636 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.09636.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Mengxi Xiao, Kailai Yang, Pengde Zhao, Enze Zhang, Ziyan Kuang, Zhiwei Liu, Weiguang Han, Shu Liao, Lianting Huang, Jinpeng Hu, Min Peng, Qianqian Xie, Sophia Ananiadou</p><p><b>Upvotes:</b> 24</p><p><b>Summary:</b> Mental health disorders affect hundreds of millions globally, and the Web now serves as a primary medium for accessing support, information, and assessment. Large language models (LLMs) offer scalable and accessible assistance, yet their deployment in mental-health settings remains risky when their reasoning is incomplete, inconsistent, or ungrounded. Existing psychological LLMs emphasize emotional understanding or knowledge recall but overlook the step-wise, clinically aligned reasoning required for appraisal, diagnosis, intervention planning, abstraction, and verification. To address these issues, we introduce MentraSuite, a unified framework for advancing reliable mental-health reasoning. We propose MentraBench, a comprehensive benchmark spanning five core reasoning aspects, six tasks, and 13 datasets, evaluating both task performance and reasoning quality across five dimensions: conciseness, coherence, hallucination avoidance, task understanding, and internal consistency. We further present Mindora, a post-trained model optimized through a hybrid SFT-RL framework with an inconsistency-detection reward to enforce faithful and coherent reasoning. To support training, we construct high-quality trajectories using a novel reasoning trajectory generation strategy, that strategically filters difficult samples and applies a structured, consistency-oriented rewriting process to produce concise, readable, and well-balanced trajectories. Across 20 evaluated LLMs, Mindora achieves the highest average performance on MentraBench and shows remarkable performances in reasoning reliability, demonstrating its effectiveness for complex mental-health scenarios.</p> + https://arxiv.org/abs/2512.09636 + Wed, 10 Dec 2025 13:26:22 +0000 - ReVSeg: Incentivizing the Reasoning Chain for Video Segmentation with Reinforcement Learning - https://arxiv.org/abs/2512.02835 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.02835.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yifan Li, Yingda Yin, Lingting Zhu, Weikai Chen, Shengju Qian, Xin Wang, Yanwei Fu</p><p><b>Upvotes:</b> 9</p><p><b>Summary:</b> Reasoning-centric video object segmentation is an inherently complex task: the query often refers to dynamics, causality, and temporal interactions, rather than static appearances. Yet existing solutions generally collapse these factors into simplified reasoning with latent embeddings, rendering the reasoning chain opaque and essentially intractable. We therefore adopt an explicit decomposition perspective and introduce ReVSeg, which executes reasoning as sequential decisions in the native interface of pretrained vision language models (VLMs). Rather than folding all reasoning into a single-step prediction, ReVSeg executes three explicit operations -- semantics interpretation, temporal evidence selection, and spatial grounding -- aligning pretrained capabilities. We further employ reinforcement learning to optimize the multi-step reasoning chain, enabling the model to self-refine its decision quality from outcome-driven signals. Experimental results demonstrate that ReVSeg attains state-of-the-art performances on standard video object segmentation benchmarks and yields interpretable reasoning trajectories. Project page is available at https://clementine24.github.io/ReVSeg/ .</p> - https://arxiv.org/abs/2512.02835 - Tue, 02 Dec 2025 14:44:12 +0000 + Openpi Comet: Competition Solution For 2025 BEHAVIOR Challenge + https://arxiv.org/abs/2512.10071 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.10071.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Junjie Bai, Yu-Wei Chao, Qizhi Chen, Jinwei Gu, Moo Jin Kim, Zhaoshuo Li, Xuan Li, Tsung-Yi Lin, Ming-Yu Liu, Nic Ma, Kaichun Mo, Delin Qu, Shangkun Sun, Hongchi Xia, Fangyin Wei, Xiaohui Zeng</p><p><b>Upvotes:</b> 17</p><p><b>Summary:</b> The 2025 BEHAVIOR Challenge is designed to rigorously track progress toward solving long-horizon tasks by physical agents in simulated environments. BEHAVIOR-1K focuses on everyday household tasks that people most want robots to assist with and these tasks introduce long-horizon mobile manipulation challenges in realistic settings, bridging the gap between current research and real-world, human-centric applications. This report presents our solution to the 2025 BEHAVIOR Challenge in a very close 2nd place and substantially outperforms the rest of the submissions. Building on π_{0.5}, we focus on systematically building our solution by studying the effects of training techniques and data. Through careful ablations, we show the scaling power in pre-training and post-training phases for competitive performance. We summarize our practical lessons and design recommendations that we hope will provide actionable insights for the broader embodied AI community when adapting powerful foundation models to complex embodied scenarios.</p> + https://arxiv.org/abs/2512.10071 + Wed, 10 Dec 2025 20:46:40 +0000 - Fast-Decoding Diffusion Language Models via Progress-Aware Confidence Schedules - https://arxiv.org/abs/2512.02892 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.02892.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Amr Mohamed, Yang Zhang, Michalis Vazirgiannis, Guokan Shang</p><p><b>Upvotes:</b> 9</p><p><b>Summary:</b> Diffusion large language models (dLLMs) offer a promising alternative to autoregressive models, but their practical utility is severely hampered by slow, iterative sampling. We present SchED, a training-free, model-agnostic early-exit algorithm that aggregates full-span logit margins and halts decoding once a smooth, progress-dependent confidence threshold is met. We evaluated SchED on two dLLM families (Dream and LLaDA), in base and instruction-tuned variants across ten benchmarks spanning downstream tasks including multiple-choice question answering (MCQ), math, long-form QA/summarization, and translation. SchED delivers large, stable accelerations: on instruction-tuned models, it achieves 3.8-4.0times speedups while retaining 99.8-100% of the baseline score on average. On base models, SchED yields consistent speedup gains with 99.1-100% performance retention, with up to 2.34times under more aggressive settings. Using a conservative speed metric that heavily penalizes quality loss (QPS, γ{=}4), we show that SchED is robust and clearly outperforms prior confidence-based early-exit methods, which break down on long-form generation. An entropy analysis of the model's token predictions reveals that instruction tuning speeds up the decay of predictive entropy. By turning genuine confidence stabilization into computational savings, SchED makes dLLM decoding substantially more efficient.</p> - https://arxiv.org/abs/2512.02892 - Tue, 02 Dec 2025 16:01:08 +0000 + Sliding Window Attention Adaptation + https://arxiv.org/abs/2512.10411 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.10411.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yijiong Yu, Jiale Liu, Qingyun Wu, Huazheng Wang, Ji Pei</p><p><b>Upvotes:</b> 17</p><p><b>Summary:</b> The self-attention mechanism in Transformer-based Large Language Models (LLMs) scales quadratically with input length, making long-context inference expensive. Sliding window attention (SWA) reduces this cost to linear complexity, but naively enabling complete SWA at inference-time for models pretrained with full attention (FA) causes severe long-context performance degradation due to training-inference mismatch. This makes us wonder: Can FA-pretrained LLMs be well adapted to SWA without pretraining? We investigate this by proposing Sliding Window Attention Adaptation (SWAA), a set of practical recipes that combine five methods for better adaptation: (1) applying SWA only during prefilling; (2) preserving "sink" tokens; (3) interleaving FA/SWA layers; (4) chain-of-thought (CoT); and (5) fine-tuning. Our experiments show that SWA adaptation is feasible while non-trivial: no single method suffices, yet specific synergistic combinations effectively recover the original long-context performance. We further analyze the performance-efficiency trade-offs of different SWAA configurations and provide recommended recipes for diverse scenarios. Our code is available at https://github.com/yuyijiong/sliding-window-attention-adaptation</p> + https://arxiv.org/abs/2512.10411 + Thu, 11 Dec 2025 08:21:24 +0000 - Self-Improving VLM Judges Without Human Annotations - https://arxiv.org/abs/2512.05145 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.05145.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Inna Wanyin Lin, Yushi Hu, Shuyue Stella Li, Scott Geng, Pang Wei Koh, Luke Zettlemoyer, Tim Althoff, Marjan Ghazvininejad</p><p><b>Upvotes:</b> 18</p><p><b>Summary:</b> Effective judges of Vision-Language Models (VLMs) are crucial for model development. Current methods for training VLM judges mainly rely on large-scale human preference annotations. However, such an approach is costly, and the annotations easily become obsolete as models rapidly improve. In this work, we present a framework to self-train a VLM judge model without any human preference annotations, using only self-synthesized data. Our method is iterative and has three stages: (1) generate diverse multimodal instruction-response pairs at varying quality levels, (2) generate reasoning traces and judgments for each pair, removing the ones that do not match our expected quality levels, and (3) training on correct judge answers and their reasoning traces. We evaluate the resulting judge on Multimodal RewardBench and VL-RewardBench across domains: correctness, preference, reasoning, safety, and visual question-answering. Our method improves a Llama-3.2-11B multimodal judge from 0.38 to 0.51 in overall accuracy on VL-RewardBench, often outperforming much larger models including Llama-3.2-90B, GPT-4o, and Claude 3.5 Sonnet, with particularly strong gains in general, hallucination, and reasoning dimensions. The overall strength of these human-annotation-free results suggest the potential for a future self-judge that evolves alongside rapidly improving VLM capabilities.</p> - https://arxiv.org/abs/2512.05145 - Tue, 02 Dec 2025 20:52:19 +0000 + CAPTAIN: Semantic Feature Injection for Memorization Mitigation in Text-to-Image Diffusion Models + https://arxiv.org/abs/2512.10655 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.10655.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Tong Zhang, Carlos Hinojosa, Bernard Ghanem</p><p><b>Upvotes:</b> 8</p><p><b>Summary:</b> Diffusion models can unintentionally reproduce training examples, raising privacy and copyright concerns as these systems are increasingly deployed at scale. Existing inference-time mitigation methods typically manipulate classifier-free guidance (CFG) or perturb prompt embeddings; however, they often struggle to reduce memorization without compromising alignment with the conditioning prompt. We introduce CAPTAIN, a training-free framework that mitigates memorization by directly modifying latent features during denoising. CAPTAIN first applies frequency-based noise initialization to reduce the tendency to replicate memorized patterns early in the denoising process. It then identifies the optimal denoising timesteps for feature injection and localizes memorized regions. Finally, CAPTAIN injects semantically aligned features from non-memorized reference images into localized latent regions, suppressing memorization while preserving prompt fidelity and visual quality. Our experiments show that CAPTAIN achieves substantial reductions in memorization compared to CFG-based baselines while maintaining strong alignment with the intended prompt.</p> + https://arxiv.org/abs/2512.10655 + Thu, 11 Dec 2025 14:01:47 +0000 - SPARK: Stepwise Process-Aware Rewards for Reference-Free Reinforcement Learning - https://arxiv.org/abs/2512.03244 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.03244.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Salman Rahman, Sruthi Gorantla, Arpit Gupta, Swastik Roy, Nanyun Peng, Yang Liu</p><p><b>Upvotes:</b> 15</p><p><b>Summary:</b> Process reward models (PRMs) that provide dense, step-level feedback have shown promise for reinforcement learning, yet their adoption remains limited by the need for expensive step-level annotations or ground truth references. We propose SPARK: a three-stage framework where in the first stage a generator model produces diverse solutions and a verifier model evaluates them using parallel scaling (self-consistency) and sequential scaling (meta-critique). In the second stage, we use these verification outputs as synthetic training data to fine-tune generative process reward models, which subsequently serve as reward signals during training. We show that aggregating multiple independent verifications at the step level produces training data for process reward models that surpass ground-truth outcome supervision, achieving 67.5 F1 on ProcessBench (a benchmark for identifying erroneous steps in mathematical reasoning) compared to 66.4 for reference-guided training and 61.9 for GPT-4o. In the final stage, we apply our generative PRM with chain-of-thought verification (PRM-CoT) as the reward model in RL experiments on mathematical reasoning, and introduce format constraints to prevent reward hacking. Using Qwen2.5-Math-7B, we achieve 47.4% average accuracy across six mathematical reasoning benchmarks, outperforming ground-truth-based RLVR (43.9%). Our work enables reference-free RL training that exceeds ground-truth methods, opening new possibilities for domains lacking verifiable answers or accessible ground truth.</p> - https://arxiv.org/abs/2512.03244 - Tue, 02 Dec 2025 21:30:47 +0000 + Sharp Monocular View Synthesis in Less Than a Second + https://arxiv.org/abs/2512.10685 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.10685.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Lars Mescheder, Wei Dong, Shiwei Li, Xuyang Bai, Marcel Santos, Peiyun Hu, Bruno Lecouat, Mingmin Zhen, Amaël Delaunoy, Tian Fang, Yanghai Tsin, Stephan R. Richter, Vladlen Koltun</p><p><b>Upvotes:</b> 7</p><p><b>Summary:</b> We present SHARP, an approach to photorealistic view synthesis from a single image. Given a single photograph, SHARP regresses the parameters of a 3D Gaussian representation of the depicted scene. This is done in less than a second on a standard GPU via a single feedforward pass through a neural network. The 3D Gaussian representation produced by SHARP can then be rendered in real time, yielding high-resolution photorealistic images for nearby views. The representation is metric, with absolute scale, supporting metric camera movements. Experimental results demonstrate that SHARP delivers robust zero-shot generalization across datasets. It sets a new state of the art on multiple datasets, reducing LPIPS by 25-34% and DISTS by 21-43% versus the best prior model, while lowering the synthesis time by three orders of magnitude. Code and weights are provided at https://github.com/apple/ml-sharp</p> + https://arxiv.org/abs/2512.10685 + Thu, 11 Dec 2025 14:34:11 +0000 - M3DR: Towards Universal Multilingual Multimodal Document Retrieval - https://arxiv.org/abs/2512.03514 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.03514.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Adithya S Kolavi, Vyoman Jain</p><p><b>Upvotes:</b> 7</p><p><b>Summary:</b> Multimodal document retrieval systems have shown strong progress in aligning visual and textual content for semantic search. However, most existing approaches remain heavily English-centric, limiting their effectiveness in multilingual contexts. In this work, we present M3DR (Multilingual Multimodal Document Retrieval), a framework designed to bridge this gap across languages, enabling applicability across diverse linguistic and cultural contexts. M3DR leverages synthetic multilingual document data and generalizes across different vision-language architectures and model sizes, enabling robust cross-lingual and cross-modal alignment. Using contrastive training, our models learn unified representations for text and document images that transfer effectively across languages. We validate this capability on 22 typologically diverse languages, demonstrating consistent performance and adaptability across linguistic and script variations. We further introduce a comprehensive benchmark that captures real-world multilingual scenarios, evaluating models under monolingual, multilingual, and mixed-language settings. M3DR generalizes across both single dense vector and ColBERT-style token-level multi-vector retrieval paradigms. Our models, NetraEmbed and ColNetraEmbed achieve state-of-the-art performance with ~150% relative improvements on cross-lingual retrieval.</p> - https://arxiv.org/abs/2512.03514 - Wed, 03 Dec 2025 07:17:59 +0000 + What matters for Representation Alignment: Global Information or Spatial Structure? + https://arxiv.org/abs/2512.10794 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.10794.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jaskirat Singh, Xingjian Leng, Zongze Wu, Liang Zheng, Richard Zhang, Eli Shechtman, Saining Xie</p><p><b>Upvotes:</b> 8</p><p><b>Summary:</b> Representation alignment (REPA) guides generative training by distilling representations from a strong, pretrained vision encoder to intermediate diffusion features. We investigate a fundamental question: what aspect of the target representation matters for generation, its global semantic information (e.g., measured by ImageNet-1K accuracy) or its spatial structure (i.e. pairwise cosine similarity between patch tokens)? Prevalent wisdom holds that stronger global semantic performance leads to better generation as a target representation. To study this, we first perform a large-scale empirical analysis across 27 different vision encoders and different model scales. The results are surprising; spatial structure, rather than global performance, drives the generation performance of a target representation. To further study this, we introduce two straightforward modifications, which specifically accentuate the transfer of spatial information. We replace the standard MLP projection layer in REPA with a simple convolution layer and introduce a spatial normalization layer for the external representation. Surprisingly, our simple method (implemented in <4 lines of code), termed iREPA, consistently improves convergence speed of REPA, across a diverse set of vision encoders, model sizes, and training variants (such as REPA, REPA-E, Meanflow, JiT etc). %, etc. Our work motivates revisiting the fundamental working mechanism of representational alignment and how it can be leveraged for improved training of generative models. The code and project page are available at https://end2end-diffusion.github.io/irepa</p> + https://arxiv.org/abs/2512.10794 + Thu, 11 Dec 2025 16:39:53 +0000 - TwinFlow: Realizing One-step Generation on Large Models with Self-adversarial Flows - https://arxiv.org/abs/2512.05150 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.05150.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zhenglin Cheng, Peng Sun, Jianguo Li, Tao Lin</p><p><b>Upvotes:</b> 68</p><p><b>Summary:</b> Recent advances in large multi-modal generative models have demonstrated impressive capabilities in multi-modal generation, including image and video generation. These models are typically built upon multi-step frameworks like diffusion and flow matching, which inherently limits their inference efficiency (requiring 40-100 Number of Function Evaluations (NFEs)). While various few-step methods aim to accelerate the inference, existing solutions have clear limitations. Prominent distillation-based methods, such as progressive and consistency distillation, either require an iterative distillation procedure or show significant degradation at very few steps (< 4-NFE). Meanwhile, integrating adversarial training into distillation (e.g., DMD/DMD2 and SANA-Sprint) to enhance performance introduces training instability, added complexity, and high GPU memory overhead due to the auxiliary trained models. To this end, we propose TwinFlow, a simple yet effective framework for training 1-step generative models that bypasses the need of fixed pretrained teacher models and avoids standard adversarial networks during training, making it ideal for building large-scale, efficient models. On text-to-image tasks, our method achieves a GenEval score of 0.83 in 1-NFE, outperforming strong baselines like SANA-Sprint (a GAN loss-based framework) and RCGM (a consistency-based framework). Notably, we demonstrate the scalability of TwinFlow by full-parameter training on Qwen-Image-20B and transform it into an efficient few-step generator. With just 1-NFE, our approach matches the performance of the original 100-NFE model on both the GenEval and DPG-Bench benchmarks, reducing computational cost by 100times with minor quality degradation. Project page is available at https://zhenglin-cheng.com/twinflow.</p> - https://arxiv.org/abs/2512.05150 - Wed, 03 Dec 2025 07:45:46 +0000 + MMSI-Video-Bench: A Holistic Benchmark for Video-Based Spatial Intelligence + https://arxiv.org/abs/2512.10863 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.10863.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jingli Lin, Runsen Xu, Shaohao Zhu, Sihan Yang, Peizhou Cao, Yunlong Ran, Miao Hu, Chenming Zhu, Yiman Xie, Yilin Long, Wenbo Hu, Dahua Lin, Tai Wang, Jiangmiao Pang</p><p><b>Upvotes:</b> 20</p><p><b>Summary:</b> Spatial understanding over continuous visual input is crucial for MLLMs to evolve into general-purpose assistants in physical environments. Yet there is still no comprehensive benchmark that holistically assesses the progress toward this goal. In this work, we introduce MMSI-Video-Bench, a fully human-annotated benchmark for video-based spatial intelligence in MLLMs. It operationalizes a four-level framework, Perception, Planning, Prediction, and Cross-Video Reasoning, through 1,106 questions grounded in 1,278 clips from 25 datasets and in-house videos. Each item is carefully designed and reviewed by 3DV experts with explanatory rationales to ensure precise, unambiguous grounding. Leveraging its diverse data sources and holistic task coverage, MMSI-Video-Bench also supports three domain-oriented sub-benchmarks (Indoor Scene Perception Bench, Robot Bench and Grounding Bench) for targeted capability assessment. We evaluate 25 strong open-source and proprietary MLLMs, revealing a striking human--AI gap: many models perform near chance, and the best reasoning model lags humans by nearly 60%. We further find that spatially fine-tuned models still fail to generalize effectively on our benchmark. Fine-grained error analysis exposes systematic failures in geometric reasoning, motion grounding, long-horizon prediction, and cross-video correspondence. We also show that typical frame-sampling strategies transfer poorly to our reasoning-intensive benchmark, and that neither 3D spatial cues nor chain-of-thought prompting yields meaningful gains. We expect our benchmark to establish a solid testbed for advancing video-based spatial intelligence.</p> + https://arxiv.org/abs/2512.10863 + Thu, 11 Dec 2025 17:57:24 +0000 - ReCamDriving: LiDAR-Free Camera-Controlled Novel Trajectory Video Generation - https://arxiv.org/abs/2512.03621 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.03621.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yaokun Li, Shuaixian Wang, Mantang Guo, Jiehui Huang, Taojun Ding, Mu Hu, Kaixuan Wang, Shaojie Shen, Guang Tan</p><p><b>Upvotes:</b> 8</p><p><b>Summary:</b> We propose ReCamDriving, a purely vision-based, camera-controlled novel-trajectory video generation framework. While repair-based methods fail to restore complex artifacts and LiDAR-based approaches rely on sparse and incomplete cues, ReCamDriving leverages dense and scene-complete 3DGS renderings for explicit geometric guidance, achieving precise camera-controllable generation. To mitigate overfitting to restoration behaviors when conditioned on 3DGS renderings, ReCamDriving adopts a two-stage training paradigm: the first stage uses camera poses for coarse control, while the second stage incorporates 3DGS renderings for fine-grained viewpoint and geometric guidance. Furthermore, we present a 3DGS-based cross-trajectory data curation strategy to eliminate the train-test gap in camera transformation patterns, enabling scalable multi-trajectory supervision from monocular videos. Based on this strategy, we construct the ParaDrive dataset, containing over 110K parallel-trajectory video pairs. Extensive experiments demonstrate that ReCamDriving achieves state-of-the-art camera controllability and structural consistency.</p> - https://arxiv.org/abs/2512.03621 - Wed, 03 Dec 2025 09:55:25 +0000 + Insight Miner: A Time Series Analysis Dataset for Cross-Domain Alignment with Natural Language + https://arxiv.org/abs/2512.11251 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.11251.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yunkai Zhang, Yawen Zhang, Ming Zheng, Kezhen Chen, Chongyang Gao, Ruian Ge, Siyuan Teng, Amine Jelloul, Jinmeng Rao, Xiaoyuan Guo, Chiang-Wei Fang, Zeyu Zheng, Jie Yang</p><p><b>Upvotes:</b> 5</p><p><b>Summary:</b> Time-series data is critical across many scientific and industrial domains, including environmental analysis, agriculture, transportation, and finance. However, mining insights from this data typically requires deep domain expertise, a process that is both time-consuming and labor-intensive. In this paper, we propose Insight Miner, a large-scale multimodal model (LMM) designed to generate high-quality, comprehensive time-series descriptions enriched with domain-specific knowledge. To facilitate this, we introduce TS-InsightsAvailable at \href{https://huggingface.co/datasets/zhykoties/time-series-language-alignment{https://huggingface.co/datasets/zhykoties/time-series-language-alignment}.}, the first general-domain dataset for time series and language alignment. TS-Insights contains 100k time-series windows sampled from 20 forecasting datasets. We construct this dataset using a novel agentic workflow, where we use statistical tools to extract features from raw time series before synthesizing them into coherent trend descriptions with GPT-4. Following instruction tuning on TS-Insights, Insight Miner outperforms state-of-the-art multimodal models, such as LLaVA liu2023llava and GPT-4, in generating time-series descriptions and insights. Our findings suggest a promising direction for leveraging LMMs in time series analysis, and serve as a foundational step toward enabling LLMs to interpret time series as a native input modality.</p> + https://arxiv.org/abs/2512.11251 + Fri, 12 Dec 2025 03:18:00 +0000 - Colon-X: Advancing Intelligent Colonoscopy from Multimodal Understanding to Clinical Reasoning - https://arxiv.org/abs/2512.03667 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.03667.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Ge-Peng Ji, Jingyi Liu, Deng-Ping Fan, Nick Barnes</p><p><b>Upvotes:</b> 3</p><p><b>Summary:</b> In this study, we present Colon-X, an open initiative aimed at advancing multimodal intelligence in colonoscopy. We begin by constructing ColonVQA, the most comprehensive multimodal dataset ever built for colonoscopy, featuring over 1.1M+ visual question answering entries across 76 clinical findings and 18 multimodal tasks. Beyond serving as a community-wide data foundation, we further investigate a critical yet underexplored transition in colonoscopy - evolving from multimodal understanding to clinical reasoning: (a) To capture the current landscape of multimodal understanding behaviors, we systematically assess the generalizability of 22 multimodal large language models and examine their reliability under human-induced perturbations. The results reveal that clinical outputs from leading MLLMs remain far from robust and trustworthy. (b) To narrow this gap, we further explore reasoning-centric intelligence tailored for colonoscopy. Specifically, we curate ColonReason, a clinically grounded reasoning dataset annotated through a multi-expert debating pipeline, and develop ColonR1, the first R1-styled model incorporating task-adaptive rewarding and gradient-stable optimization techniques. Under data-scarce conditions, our ColonR1 achieves 56.61% overall accuracy, outperforming supervised fine-tuning by 25.22%, and sets a new reasoning-enabled baseline for multimodal colonoscopy analysis. All data and model resources are publicly available at https://github.com/ai4colonoscopy/Colon-X.</p> - https://arxiv.org/abs/2512.03667 - Wed, 03 Dec 2025 10:55:07 +0000 + PersonaLive! Expressive Portrait Image Animation for Live Streaming + https://arxiv.org/abs/2512.11253 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.11253.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zhiyuan Li, Chi-Man Pun, Chen Fang, Jue Wang, Xiaodong Cun</p><p><b>Upvotes:</b> 27</p><p><b>Summary:</b> Current diffusion-based portrait animation models predominantly focus on enhancing visual quality and expression realism, while overlooking generation latency and real-time performance, which restricts their application range in the live streaming scenario. We propose PersonaLive, a novel diffusion-based framework towards streaming real-time portrait animation with multi-stage training recipes. Specifically, we first adopt hybrid implicit signals, namely implicit facial representations and 3D implicit keypoints, to achieve expressive image-level motion control. Then, a fewer-step appearance distillation strategy is proposed to eliminate appearance redundancy in the denoising process, greatly improving inference efficiency. Finally, we introduce an autoregressive micro-chunk streaming generation paradigm equipped with a sliding training strategy and a historical keyframe mechanism to enable low-latency and stable long-term video generation. Extensive experiments demonstrate that PersonaLive achieves state-of-the-art performance with up to 7-22x speedup over prior diffusion-based portrait animation models.</p> + https://arxiv.org/abs/2512.11253 + Fri, 12 Dec 2025 03:24:40 +0000 - VideoSSM: Autoregressive Long Video Generation with Hybrid State-Space Memory - https://arxiv.org/abs/2512.04519 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.04519.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yifei Yu, Xiaoshan Wu, Xinting Hu, Tao Hu, Yangtian Sun, Xiaoyang Lyu, Bo Wang, Lin Ma, Yuewen Ma, Zhongrui Wang, Xiaojuan Qi</p><p><b>Upvotes:</b> 3</p><p><b>Summary:</b> Autoregressive (AR) diffusion enables streaming, interactive long-video generation by producing frames causally, yet maintaining coherence over minute-scale horizons remains challenging due to accumulated errors, motion drift, and content repetition. We approach this problem from a memory perspective, treating video synthesis as a recurrent dynamical process that requires coordinated short- and long-term context. We propose VideoSSM, a Long Video Model that unifies AR diffusion with a hybrid state-space memory. The state-space model (SSM) serves as an evolving global memory of scene dynamics across the entire sequence, while a context window provides local memory for motion cues and fine details. This hybrid design preserves global consistency without frozen, repetitive patterns, supports prompt-adaptive interaction, and scales in linear time with sequence length. Experiments on short- and long-range benchmarks demonstrate state-of-the-art temporal consistency and motion stability among autoregressive video generator especially at minute-scale horizons, enabling content diversity and interactive prompt-based control, thereby establishing a scalable, memory-aware framework for long video generation.</p> - https://arxiv.org/abs/2512.04519 - Thu, 04 Dec 2025 07:06:02 +0000 + Exploring MLLM-Diffusion Information Transfer with MetaCanvas + https://arxiv.org/abs/2512.11464 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.11464.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Han Lin, Xichen Pan, Ziqi Huang, Ji Hou, Jialiang Wang, Weifeng Chen, Zecheng He, Felix Juefei-Xu, Junzhe Sun, Zhipeng Fan, Ali Thabet, Mohit Bansal, Chu Wang</p><p><b>Upvotes:</b> 12</p><p><b>Summary:</b> Multimodal learning has rapidly advanced visual understanding, largely via multimodal large language models (MLLMs) that use powerful LLMs as cognitive cores. In visual generation, however, these powerful core models are typically reduced to global text encoders for diffusion models, leaving most of their reasoning and planning ability unused. This creates a gap: current multimodal LLMs can parse complex layouts, attributes, and knowledge-intensive scenes, yet struggle to generate images or videos with equally precise and structured control. We propose MetaCanvas, a lightweight framework that lets MLLMs reason and plan directly in spatial and spatiotemporal latent spaces and interface tightly with diffusion generators. We empirically implement MetaCanvas on three different diffusion backbones and evaluate it across six tasks, including text-to-image generation, text/image-to-video generation, image/video editing, and in-context video generation, each requiring precise layouts, robust attribute binding, and reasoning-intensive control. MetaCanvas consistently outperforms global-conditioning baselines, suggesting that treating MLLMs as latent-space planners is a promising direction for narrowing the gap between multimodal understanding and generation.</p> + https://arxiv.org/abs/2512.11464 + Fri, 12 Dec 2025 11:07:11 +0000 - X-Humanoid: Robotize Human Videos to Generate Humanoid Videos at Scale - https://arxiv.org/abs/2512.04537 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.04537.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Pei Yang, Hai Ci, Yiren Song, Mike Zheng Shou</p><p><b>Upvotes:</b> 4</p><p><b>Summary:</b> The advancement of embodied AI has unlocked significant potential for intelligent humanoid robots. However, progress in both Vision-Language-Action (VLA) models and world models is severely hampered by the scarcity of large-scale, diverse training data. A promising solution is to "robotize" web-scale human videos, which has been proven effective for policy training. However, these solutions mainly "overlay" robot arms to egocentric videos, which cannot handle complex full-body motions and scene occlusions in third-person videos, making them unsuitable for robotizing humans. To bridge this gap, we introduce X-Humanoid, a generative video editing approach that adapts the powerful Wan 2.2 model into a video-to-video structure and finetunes it for the human-to-humanoid translation task. This finetuning requires paired human-humanoid videos, so we designed a scalable data creation pipeline, turning community assets into 17+ hours of paired synthetic videos using Unreal Engine. We then apply our trained model to 60 hours of the Ego-Exo4D videos, generating and releasing a new large-scale dataset of over 3.6 million "robotized" humanoid video frames. Quantitative analysis and user studies confirm our method's superiority over existing baselines: 69% of users rated it best for motion consistency, and 62.1% for embodiment correctness.</p> - https://arxiv.org/abs/2512.04537 - Thu, 04 Dec 2025 07:34:08 +0000 + Rethinking Expert Trajectory Utilization in LLM Post-training + https://arxiv.org/abs/2512.11470 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.11470.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Bowen Ding, Yuhan Chen, Jiayang Lv, Jiyao Yuan, Qi Zhu, Shuangshuang Tian, Dantong Zhu, Futing Wang, Heyuan Deng, Fei Mi, Lifeng Shang, Tao Lin</p><p><b>Upvotes:</b> 6</p><p><b>Summary:</b> While effective post-training integrates Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL), the optimal mechanism for utilizing expert trajectories remains unresolved. We propose the Plasticity-Ceiling Framework to theoretically ground this landscape, decomposing performance into foundational SFT performance and the subsequent RL plasticity. Through extensive benchmarking, we establish the Sequential SFT-then-RL pipeline as the superior standard, overcoming the stability deficits of synchronized approaches. Furthermore, we derive precise scaling guidelines: (1) Transitioning to RL at the SFT Stable or Mild Overfitting Sub-phase maximizes the final ceiling by securing foundational SFT performance without compromising RL plasticity; (2) Refuting ``Less is More'' in the context of SFT-then-RL scaling, we demonstrate that Data Scale determines the primary post-training potential, while Trajectory Difficulty acts as a performance multiplier; and (3) Identifying that the Minimum SFT Validation Loss serves as a robust indicator for selecting the expert trajectories that maximize the final performance ceiling. Our findings provide actionable guidelines for maximizing the value extracted from expert trajectories.</p> + https://arxiv.org/abs/2512.11470 + Fri, 12 Dec 2025 11:13:00 +0000 - COOPER: A Unified Model for Cooperative Perception and Reasoning in Spatial Intelligence - https://arxiv.org/abs/2512.04563 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.04563.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zefeng Zhang, Xiangzhao Hao, Hengzhu Tang, Zhenyu Zhang, Jiawei Sheng, Xiaodong Li, Zhenyang Li, Li Gao, Daiting Shi, Dawei Yin, Tingwen Liu</p><p><b>Upvotes:</b> 13</p><p><b>Summary:</b> Visual Spatial Reasoning is crucial for enabling Multimodal Large Language Models (MLLMs) to understand object properties and spatial relationships, yet current models still struggle with 3D-aware reasoning. Existing approaches typically enhance either perception, by augmenting RGB inputs with auxiliary modalities such as depth and segmentation, or reasoning, by training on spatial VQA datasets and applying reinforcement learning, and thus treat these two aspects in isolation. In this work, we investigate whether a unified MLLM can develop an intrinsic ability to enhance spatial perception and, through adaptive interleaved reasoning, achieve stronger spatial intelligence. We propose COOPER, a unified MLLM that leverages depth and segmentation as auxiliary modalities and is trained in two stages to acquire auxiliary modality generation and adaptive, interleaved reasoning capabilities. COOPER achieves an average 6.91\% improvement in spatial reasoning while maintaining general performance. Moreover, even a variant trained only for auxiliary modality generation attains a 7.92\% gain on distance and size estimation, suggesting that learning to generate auxiliary modalities helps internalize spatial knowledge and strengthen spatial understanding.</p> - https://arxiv.org/abs/2512.04563 - Thu, 04 Dec 2025 08:26:04 +0000 + DentalGPT: Incentivizing Multimodal Complex Reasoning in Dentistry + https://arxiv.org/abs/2512.11558 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.11558.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zhenyang Cai, Jiaming Zhang, Junjie Zhao, Ziyi Zeng, Yanchao Li, Jingyi Liang, Junying Chen, Yunjin Yang, Jiajun You, Shuzhi Deng, Tongfei Wang, Wanting Chen, Chunxiu Hao, Ruiqi Xie, Zhenwei Wen, Xiangyi Feng, Zou Ting, Jin Zou Lin, Jianquan Li, Guangjun Yu, Liangyi Chen, Junwen Wang, Shan Jiang, Benyou Wang</p><p><b>Upvotes:</b> 41</p><p><b>Summary:</b> Reliable interpretation of multimodal data in dentistry is essential for automated oral healthcare, yet current multimodal large language models (MLLMs) struggle to capture fine-grained dental visual details and lack sufficient reasoning ability for precise diagnosis. To address these limitations, we present DentalGPT, a specialized dental MLLM developed through high-quality domain knowledge injection and reinforcement learning. Specifically, the largest annotated multimodal dataset for dentistry to date was constructed by aggregating over 120k dental images paired with detailed descriptions that highlight diagnostically relevant visual features, making it the multimodal dataset with the most extensive collection of dental images to date. Training on this dataset significantly enhances the MLLM's visual understanding of dental conditions, while the subsequent reinforcement learning stage further strengthens its capability for multimodal complex reasoning. Comprehensive evaluations on intraoral and panoramic benchmarks, along with dental subsets of medical VQA benchmarks, show that DentalGPT achieves superior performance in disease classification and dental VQA tasks, outperforming many state-of-the-art MLLMs despite having only 7B parameters. These results demonstrate that high-quality dental data combined with staged adaptation provides an effective pathway for building capable and domain-specialized dental MLLMs.</p> + https://arxiv.org/abs/2512.11558 + Fri, 12 Dec 2025 13:42:57 +0000 - EtCon: Edit-then-Consolidate for Reliable Knowledge Editing - https://arxiv.org/abs/2512.04753 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.04753.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Ruilin Li, Yibin Wang, Wenhong Zhu, Chenglin Li, Jinghao Zhang, Chenliang Li, Junchi Yan, Jiaqi Wang</p><p><b>Upvotes:</b> 7</p><p><b>Summary:</b> Knowledge editing aims to update specific facts in large language models (LLMs) without full retraining. Prior efforts sought to tune the knowledge layers of LLMs, proving effective for making selective edits. However, a significant gap exists between their performance in controlled, teacher-forcing evaluations and their real-world effectiveness in lifelong learning scenarios, which greatly limits their practical applicability. This work's empirical analysis reveals two recurring issues associated with this gap: (1) Most traditional methods lead the edited model to overfit to the new fact, thereby degrading pre-trained capabilities; (2) There is a critical absence of a knowledge consolidation stage, leaving new facts insufficiently integrated into LLMs' inference-time behavior under autoregressive generation, thereby leading to a mismatch between parametric knowledge and actual generation behavior. To this end, we propose Edit-then-Consolidate, a novel knowledge editing paradigm that aims to bridge the gap between theoretical knowledge editing methods and their real-world applicability. Specifically, (1) our framework mitigates overfitting via Targeted Proximal Supervised Fine-Tuning (TPSFT) that localizes the edit via a trust-region objective to limit policy drift; (2) Then, a consolidation stage using Group Relative Policy Optimization (GRPO) aligns the edited knowledge with CoT-based inference policy by optimizing trajectory-level behavior under comprehensive reward signals. Extensive experiments demonstrate our framework consistently improves editing reliability and generalization under real-world evaluations, while better preserving locality and pre-trained capabilities.</p> - https://arxiv.org/abs/2512.04753 - Thu, 04 Dec 2025 12:43:50 +0000 + SVG-T2I: Scaling Up Text-to-Image Latent Diffusion Model Without Variational Autoencoder + https://arxiv.org/abs/2512.11749 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.11749.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Minglei Shi, Haolin Wang, Borui Zhang, Wenzhao Zheng, Bohan Zeng, Ziyang Yuan, Xiaoshi Wu, Yuanxing Zhang, Huan Yang, Xintao Wang, Pengfei Wan, Kun Gai, Jie Zhou, Jiwen Lu</p><p><b>Upvotes:</b> 34</p><p><b>Summary:</b> Visual generation grounded in Visual Foundation Model (VFM) representations offers a highly promising unified pathway for integrating visual understanding, perception, and generation. Despite this potential, training large-scale text-to-image diffusion models entirely within the VFM representation space remains largely unexplored. To bridge this gap, we scale the SVG (Self-supervised representations for Visual Generation) framework, proposing SVG-T2I to support high-quality text-to-image synthesis directly in the VFM feature domain. By leveraging a standard text-to-image diffusion pipeline, SVG-T2I achieves competitive performance, reaching 0.75 on GenEval and 85.78 on DPG-Bench. This performance validates the intrinsic representational power of VFMs for generative tasks. We fully open-source the project, including the autoencoder and generation model, together with their training, inference, evaluation pipelines, and pre-trained weights, to facilitate further research in representation-driven visual generation.</p> + https://arxiv.org/abs/2512.11749 + Fri, 12 Dec 2025 17:45:03 +0000 - MemLoRA: Distilling Expert Adapters for On-Device Memory Systems - https://arxiv.org/abs/2512.04763 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.04763.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Massimo Bini, Ondrej Bohdal, Umberto Michieli, Zeynep Akata, Mete Ozay, Taha Ceritli</p><p><b>Upvotes:</b> 3</p><p><b>Summary:</b> Memory-augmented Large Language Models (LLMs) have demonstrated remarkable consistency during prolonged dialogues by storing relevant memories and incorporating them as context. Such memory-based personalization is also key in on-device settings that allow users to keep their conversations and data private. However, memory-augmented systems typically rely on LLMs that are too costly for local on-device deployment. Even though Small Language Models (SLMs) are more suitable for on-device inference than LLMs, they cannot achieve sufficient performance. Additionally, these LLM-based systems lack native visual capabilities, limiting their applicability in multimodal contexts. In this paper, we introduce (i) MemLoRA, a novel memory system that enables local deployment by equipping SLMs with specialized memory adapters, and (ii) its vision extension MemLoRA-V, which integrates small Vision-Language Models (SVLMs) to memory systems, enabling native visual understanding. Following knowledge distillation principles, each adapter is trained separately for specific memory operationsx2013knowledge extraction, memory update, and memory-augmented generation. Equipped with memory adapters, small models enable accurate on-device memory operations without cloud dependency. On text-only operations, MemLoRA outperforms 10times larger baseline models (e.g., Gemma2-27B) and achieves performance comparable to 60times larger models (e.g., GPT-OSS-120B) on the LoCoMo benchmark. To evaluate visual understanding operations instead, we extend LoCoMo with challenging Visual Question Answering tasks that require direct visual reasoning. On this, our VLM-integrated MemLoRA-V shows massive improvements over caption-based approaches (81.3 vs. 23.7 accuracy) while keeping strong performance in text-based tasks, demonstrating the efficacy of our method in multimodal contexts.</p> - https://arxiv.org/abs/2512.04763 - Thu, 04 Dec 2025 12:56:30 +0000 + Structure From Tracking: Distilling Structure-Preserving Motion for Video Generation + https://arxiv.org/abs/2512.11792 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.11792.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yang Fei, George Stoica, Jingyuan Liu, Qifeng Chen, Ranjay Krishna, Xiaojuan Wang, Benlin Liu</p><p><b>Upvotes:</b> 9</p><p><b>Summary:</b> Reality is a dance between rigid constraints and deformable structures. For video models, that means generating motion that preserves fidelity as well as structure. Despite progress in diffusion models, producing realistic structure-preserving motion remains challenging, especially for articulated and deformable objects such as humans and animals. Scaling training data alone, so far, has failed to resolve physically implausible transitions. Existing approaches rely on conditioning with noisy motion representations, such as optical flow or skeletons extracted using an external imperfect model. To address these challenges, we introduce an algorithm to distill structure-preserving motion priors from an autoregressive video tracking model (SAM2) into a bidirectional video diffusion model (CogVideoX). With our method, we train SAM2VideoX, which contains two innovations: (1) a bidirectional feature fusion module that extracts global structure-preserving motion priors from a recurrent model like SAM2; (2) a Local Gram Flow loss that aligns how local features move together. Experiments on VBench and in human studies show that SAM2VideoX delivers consistent gains (+2.60\% on VBench, 21-22\% lower FVD, and 71.4\% human preference) over prior baselines. Specifically, on VBench, we achieve 95.51\%, surpassing REPA (92.91\%) by 2.60\%, and reduce FVD to 360.57, a 21.20\% and 22.46\% improvement over REPA- and LoRA-finetuning, respectively. The project website can be found at https://sam2videox.github.io/ .</p> + https://arxiv.org/abs/2512.11792 + Fri, 12 Dec 2025 18:56:35 +0000 - EMMA: Efficient Multimodal Understanding, Generation, and Editing with a Unified Architecture - https://arxiv.org/abs/2512.04810 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.04810.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Xin He, Longhui Wei, Jianbo Ouyang, Lingxi Xie, Qi Tian</p><p><b>Upvotes:</b> 25</p><p><b>Summary:</b> We propose EMMA, an efficient and unified architecture for multimodal understanding, generation and editing. Specifically, EMMA primarily consists of 1) An efficient autoencoder with a 32x compression ratio, which significantly reduces the number of tokens required for generation. This also ensures the training balance between understanding and generation tasks by applying the same compression ratio to images. 2) Channel-wise concatenation instead of token-wise concatenation among visual understanding and generation tokens, which further reduces the visual tokens in unified architectures. 3) A shared-and-decoupled network that enables mutual improvements across tasks while meeting the task-specific modeling requirements. 4) A mixture-of-experts mechanism adopted for visual understanding encoder, which substantially improves perceptual capabilities with a few parameters increase. Extensive experiments have shown that EMMA-4B can significantly outperform state-of-the-art unified multimodal approaches (e.g., BAGEL-7B) in both efficiency and performance, while also achieving competitive results compared to recent multimodal understanding and generation experts (e.g., Qwen3-VL and Qwen-Image). We believe that EMMA lays a solid foundation for the future development of unified multimodal architectures.</p> - https://arxiv.org/abs/2512.04810 - Thu, 04 Dec 2025 14:01:53 +0000 + V-RGBX: Video Editing with Accurate Controls over Intrinsic Properties + https://arxiv.org/abs/2512.11799 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.11799.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Ye Fang, Tong Wu, Valentin Deschaintre, Duygu Ceylan, Iliyan Georgiev, Chun-Hao Paul Huang, Yiwei Hu, Xuelin Chen, Tuanfeng Yang Wang</p><p><b>Upvotes:</b> 29</p><p><b>Summary:</b> Large-scale video generation models have shown remarkable potential in modeling photorealistic appearance and lighting interactions in real-world scenes. However, a closed-loop framework that jointly understands intrinsic scene properties (e.g., albedo, normal, material, and irradiance), leverages them for video synthesis, and supports editable intrinsic representations remains unexplored. We present V-RGBX, the first end-to-end framework for intrinsic-aware video editing. V-RGBX unifies three key capabilities: (1) video inverse rendering into intrinsic channels, (2) photorealistic video synthesis from these intrinsic representations, and (3) keyframe-based video editing conditioned on intrinsic channels. At the core of V-RGBX is an interleaved conditioning mechanism that enables intuitive, physically grounded video editing through user-selected keyframes, supporting flexible manipulation of any intrinsic modality. Extensive qualitative and quantitative results show that V-RGBX produces temporally consistent, photorealistic videos while propagating keyframe edits across sequences in a physically plausible manner. We demonstrate its effectiveness in diverse applications, including object appearance editing and scene-level relighting, surpassing the performance of prior methods.</p> + https://arxiv.org/abs/2512.11799 + Fri, 12 Dec 2025 18:59:54 +0000 - Arbitrage: Efficient Reasoning via Advantage-Aware Speculation - https://arxiv.org/abs/2512.05033 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.05033.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Monishwaran Maheswaran, Rishabh Tiwari, Yuezhou Hu, Kerem Dilmen, Coleman Hooper, Haocheng Xi, Nicholas Lee, Mehrdad Farajtabar, Michael W. Mahoney, Kurt Keutzer, Amir Gholami</p><p><b>Upvotes:</b> 15</p><p><b>Summary:</b> Modern Large Language Models achieve impressive reasoning capabilities with long Chain of Thoughts, but they incur substantial computational cost during inference, and this motivates techniques to improve the performance-cost ratio. Among these techniques, Speculative Decoding accelerates inference by employing a fast but inaccurate draft model to autoregressively propose tokens, which are then verified in parallel by a more capable target model. However, due to unnecessary rejections caused by token mismatches in semantically equivalent steps, traditional token-level Speculative Decoding struggles in reasoning tasks. Although recent works have shifted to step-level semantic verification, which improve efficiency by accepting or rejecting entire reasoning steps, existing step-level methods still regenerate many rejected steps with little improvement, wasting valuable target compute. To address this challenge, we propose Arbitrage, a novel step-level speculative generation framework that routes generation dynamically based on the relative advantage between draft and target models. Instead of applying a fixed acceptance threshold, Arbitrage uses a lightweight router trained to predict when the target model is likely to produce a meaningfully better step. This routing approximates an ideal Arbitrage Oracle that always chooses the higher-quality step, achieving near-optimal efficiency-accuracy trade-offs. Across multiple mathematical reasoning benchmarks, Arbitrage consistently surpasses prior step-level Speculative Decoding baselines, reducing inference latency by up to sim2times at matched accuracy.</p> - https://arxiv.org/abs/2512.05033 - Thu, 04 Dec 2025 17:50:53 +0000 + V-REX: Benchmarking Exploratory Visual Reasoning via Chain-of-Questions + https://arxiv.org/abs/2512.11995 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.11995.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Chenrui Fan, Yijun Liang, Shweta Bhardwaj, Kwesi Cobbina, Ming Li, Tianyi Zhou</p><p><b>Upvotes:</b> 9</p><p><b>Summary:</b> While many vision-language models (VLMs) are developed to answer well-defined, straightforward questions with highly specified targets, as in most benchmarks, they often struggle in practice with complex open-ended tasks, which usually require multiple rounds of exploration and reasoning in the visual space. Such visual thinking paths not only provide step-by-step exploration and verification as an AI detective but also produce better interpretations of the final answers. However, these paths are challenging to evaluate due to the large exploration space of intermediate steps. To bridge the gap, we develop an evaluation suite, ``Visual Reasoning with multi-step EXploration (V-REX)'', which is composed of a benchmark of challenging visual reasoning tasks requiring native multi-step exploration and an evaluation protocol. V-REX covers rich application scenarios across diverse domains. V-REX casts the multi-step exploratory reasoning into a Chain-of-Questions (CoQ) and disentangles VLMs' capability to (1) Planning: breaking down an open-ended task by selecting a chain of exploratory questions; and (2) Following: answering curated CoQ sequentially to collect information for deriving the final answer. By curating finite options of questions and answers per step, V-REX achieves a reliable quantitative and fine-grained analysis of the intermediate steps. By assessing SOTA proprietary and open-sourced VLMs, we reveal consistent scaling trends, significant differences between planning and following abilities, and substantial room for improvement in multi-step exploratory reasoning.</p> + https://arxiv.org/abs/2512.11995 + Fri, 12 Dec 2025 19:18:41 +0000 - Joint 3D Geometry Reconstruction and Motion Generation for 4D Synthesis from a Single Image - https://arxiv.org/abs/2512.05044 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.05044.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yanran Zhang, Ziyi Wang, Wenzhao Zheng, Zheng Zhu, Jie Zhou, Jiwen Lu</p><p><b>Upvotes:</b> 15</p><p><b>Summary:</b> Generating interactive and dynamic 4D scenes from a single static image remains a core challenge. Most existing generate-then-reconstruct and reconstruct-then-generate methods decouple geometry from motion, causing spatiotemporal inconsistencies and poor generalization. To address these, we extend the reconstruct-then-generate framework to jointly perform Motion generation and geometric Reconstruction for 4D Synthesis (MoRe4D). We first introduce TrajScene-60K, a large-scale dataset of 60,000 video samples with dense point trajectories, addressing the scarcity of high-quality 4D scene data. Based on this, we propose a diffusion-based 4D Scene Trajectory Generator (4D-STraG) to jointly generate geometrically consistent and motion-plausible 4D point trajectories. To leverage single-view priors, we design a depth-guided motion normalization strategy and a motion-aware module for effective geometry and dynamics integration. We then propose a 4D View Synthesis Module (4D-ViSM) to render videos with arbitrary camera trajectories from 4D point track representations. Experiments show that MoRe4D generates high-quality 4D scenes with multi-view consistency and rich dynamic details from a single image. Code: https://github.com/Zhangyr2022/MoRe4D.</p> - https://arxiv.org/abs/2512.05044 - Thu, 04 Dec 2025 17:59:10 +0000 + VOYAGER: A Training Free Approach for Generating Diverse Datasets using LLMs + https://arxiv.org/abs/2512.12072 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.12072.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Avinash Amballa, Yashas Malur Saidutta, Chi-Heng Lin, Vivek Kulkarni, Srinivas Chappidi</p><p><b>Upvotes:</b> 17</p><p><b>Summary:</b> Large language models (LLMs) are increasingly being used to generate synthetic datasets for the evaluation and training of downstream models. However, prior work has noted that such generated data lacks diversity. In this paper, we propose Voyager, a novel principled approach to generate diverse datasets. Our approach is iterative and directly optimizes a mathematical quantity that optimizes the diversity of the dataset using the machinery of determinantal point processes. Furthermore, our approach is training-free, applicable to closed-source models, and scalable. In addition to providing theoretical justification for the working of our method, we also demonstrate through comprehensive experiments that Voyager significantly outperforms popular baseline approaches by providing a 1.5-3x improvement in diversity.</p> + https://arxiv.org/abs/2512.12072 + Fri, 12 Dec 2025 22:39:01 +0000 - From Segments to Scenes: Temporal Understanding in Autonomous Driving via Vision-Language Model - https://arxiv.org/abs/2512.05277 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.05277.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Kevin Cannons, Saeed Ranjbar Alvar, Mohammad Asiful Hossain, Ahmad Rezaei, Mohsen Gholami, Alireza Heidarikhazaei, Zhou Weimin, Yong Zhang, Mohammad Akbari</p><p><b>Upvotes:</b> 4</p><p><b>Summary:</b> Temporal understanding in autonomous driving (AD) remains a significant challenge, even for recent state-of-the-art (SoTA) Vision-Language Models (VLMs). Prior work has introduced datasets and benchmarks aimed at improving temporal reasoning, but these have emphasized other video content, including sports, cooking, and movies. No existing benchmark focuses exclusively on the unique challenges of temporal understanding in ego-centric AD footage. To fill this gap, the Temporal Understanding in Autonomous Driving (TAD) benchmark is presented, which evaluates VLMs' ability to capture the dynamic relationships between actions in AD. TAD comprises nearly 6,000 question-answer (QA) pairs, spanning 7 human-designed tasks. In addition, an evaluation is performed that consists of 9 closed- and open-source generalist models as well as SoTA AD specialist models. When applied to TAD, current SoTA models demonstrated substandard accuracies, largely due to imperfect fine-grained motion understanding. To improve motion understanding and overall accuracy on TAD, two novel training-free solutions are proposed: Scene-CoT, that leverages Chain-of-Thought (CoT) and TCogMap, which incorporates an ego-centric temporal cognitive map. The proposed approaches are integrated with existing VLMs and improve average accuracy on TAD by up to 17.72%. By introducing TAD, benchmarking multiple SoTA models, and proposing effective enhancements, this work aims to catalyze future research on temporal understanding in AD. The benchmark and evaluation code are available at https://huggingface.co/datasets/vbdai/TAD{Hugging Face} and https://github.com/vbdi/tad_bench{Github}, respectively.</p> - https://arxiv.org/abs/2512.05277 - Thu, 04 Dec 2025 21:57:10 +0000 + Error-Free Linear Attention is a Free Lunch: Exact Solution from Continuous-Time Dynamics + https://arxiv.org/abs/2512.12602 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.12602.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jingdi Lei, Di Zhang, Soujanya Poria</p><p><b>Upvotes:</b> 37</p><p><b>Summary:</b> Linear-time attention and State Space Models (SSMs) promise to solve the quadratic cost bottleneck in long-context language models employing softmax attention. We introduce Error-Free Linear Attention (EFLA), a numerically stable, fully parallelism and generalized formulation of the delta rule. Specifically, we formulate the online learning update as a continuous-time dynamical system and prove that its exact solution is not only attainable but also computable in linear time with full parallelism. By leveraging the rank-1 structure of the dynamics matrix, we directly derive the exact closed-form solution effectively corresponding to the infinite-order Runge-Kutta method. This attention mechanism is theoretically free from error accumulation, perfectly capturing the continuous dynamics while preserving the linear-time complexity. Through an extensive suite of experiments, we show that EFLA enables robust performance in noisy environments, achieving lower language modeling perplexity and superior downstream benchmark performance than DeltaNet without introducing additional parameters. Our work provides a new theoretical foundation for building high-fidelity, scalable linear-time attention models.</p> + https://arxiv.org/abs/2512.12602 + Sun, 14 Dec 2025 08:51:02 +0000 - LYNX: Learning Dynamic Exits for Confidence-Controlled Reasoning - https://arxiv.org/abs/2512.05325 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.05325.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Ömer Faruk Akgül, Yusuf Hakan Kalaycı, Rajgopal Kannan, Willie Neiswanger, Viktor Prasanna</p><p><b>Upvotes:</b> 2</p><p><b>Summary:</b> Large reasoning models achieve strong performance on complex tasks by generating extended chains of thought, but they often "overthink": continuing to reason long after they have enough information to answer correctly. This wastes inference-time compute and can hurt accuracy. Existing attempts to stop early either manipulate decoding with extra sampling and heuristics, rely on auxiliary verifier models, or operate only as post-hoc analysis pipelines without formal guarantees. We introduce LYNX, an online early-exit mechanism that turns a model's own hidden-state awareness into confidence-controlled stopping decisions. LYNX attaches exit decisions to naturally occurring reasoning cues (e.g., "hmm", "wait") during generation, trains a lightweight probe on hidden states at those cue tokens using supervision from forced exits, and wraps the resulting scores in split conformal prediction to obtain distribution-free control over premature exits. Crucially, we train and calibrate this probe once on a generic mathematical corpus and reuse it unchanged across benchmarks, decoding temperatures, and even non-mathematical tasks. Across three model families spanning 1.5B to 32B parameters, a single mathematically trained probe per base model yields strong accuracy--efficiency tradeoffs. On GSM8K, LYNX matches or improves baseline accuracy while reducing tokens by 40--65\%; on MATH-500 it improves accuracy by up to 12 points with roughly 35--60\% fewer tokens; on AIME 2024 it recovers baseline accuracy with more than 50\% token savings; and on CommonsenseQA, a non-math benchmark, it transfers zero-shot with modest accuracy gains and up to 70\% fewer tokens. Compared to state-of-the-art early-exit methods, LYNX offers competitive or superior Pareto frontiers while remaining fully online, requiring no proxy models at inference, and providing explicit, user-tunable confidence guarantees.</p> - https://arxiv.org/abs/2512.05325 - Fri, 05 Dec 2025 00:04:42 +0000 + Scone: Bridging Composition and Distinction in Subject-Driven Image Generation via Unified Understanding-Generation Modeling + https://arxiv.org/abs/2512.12675 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.12675.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yuran Wang, Bohan Zeng, Chengzhuo Tong, Wenxuan Liu, Yang Shi, Xiaochen Ma, Hao Liang, Yuanxing Zhang, Wentao Zhang</p><p><b>Upvotes:</b> 40</p><p><b>Summary:</b> Subject-driven image generation has advanced from single- to multi-subject composition, while neglecting distinction, the ability to identify and generate the correct subject when inputs contain multiple candidates. This limitation restricts effectiveness in complex, realistic visual settings. We propose Scone, a unified understanding-generation method that integrates composition and distinction. Scone enables the understanding expert to act as a semantic bridge, conveying semantic information and guiding the generation expert to preserve subject identity while minimizing interference. A two-stage training scheme first learns composition, then enhances distinction through semantic alignment and attention-based masking. We also introduce SconeEval, a benchmark for evaluating both composition and distinction across diverse scenarios. Experiments demonstrate that Scone outperforms existing open-source models in composition and distinction tasks on two benchmarks. Our model, benchmark, and training data are available at: https://github.com/Ryann-Ran/Scone.</p> + https://arxiv.org/abs/2512.12675 + Sun, 14 Dec 2025 12:58:19 +0000 - SpaceControl: Introducing Test-Time Spatial Control to 3D Generative Modeling - https://arxiv.org/abs/2512.05343 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.05343.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Elisabetta Fedele, Francis Engelmann, Ian Huang, Or Litany, Marc Pollefeys, Leonidas Guibas</p><p><b>Upvotes:</b> 23</p><p><b>Summary:</b> Generative methods for 3D assets have recently achieved remarkable progress, yet providing intuitive and precise control over the object geometry remains a key challenge. Existing approaches predominantly rely on text or image prompts, which often fall short in geometric specificity: language can be ambiguous, and images are cumbersome to edit. In this work, we introduce SpaceControl, a training-free test-time method for explicit spatial control of 3D generation. Our approach accepts a wide range of geometric inputs, from coarse primitives to detailed meshes, and integrates seamlessly with modern pre-trained generative models without requiring any additional training. A controllable parameter lets users trade off between geometric fidelity and output realism. Extensive quantitative evaluation and user studies demonstrate that SpaceControl outperforms both training-based and optimization-based baselines in geometric faithfulness while preserving high visual quality. Finally, we present an interactive user interface that enables online editing of superquadrics for direct conversion into textured 3D assets, facilitating practical deployment in creative workflows. Find our project page at https://spacecontrol3d.github.io/</p> - https://arxiv.org/abs/2512.05343 - Fri, 05 Dec 2025 00:54:48 +0000 + WebOperator: Action-Aware Tree Search for Autonomous Agents in Web Environment + https://arxiv.org/abs/2512.12692 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.12692.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Mahir Labib Dihan, Tanzima Hashem, Mohammed Eunus Ali, Md Rizwan Parvez</p><p><b>Upvotes:</b> 13</p><p><b>Summary:</b> LLM-based agents often operate in a greedy, step-by-step manner, selecting actions solely based on the current observation without considering long-term consequences or alternative paths. This lack of foresight is particularly problematic in web environments, which are only partially observable-limited to browser-visible content (e.g., DOM and UI elements)-where a single misstep often requires complex and brittle navigation to undo. Without an explicit backtracking mechanism, agents struggle to correct errors or systematically explore alternative paths. Tree-search methods provide a principled framework for such structured exploration, but existing approaches lack mechanisms for safe backtracking, making them prone to unintended side effects. They also assume that all actions are reversible, ignoring the presence of irreversible actions-limitations that reduce their effectiveness in realistic web tasks. To address these challenges, we introduce WebOperator, a tree-search framework that enables reliable backtracking and strategic exploration. Our method incorporates a best-first search strategy that ranks actions by both reward estimates and safety considerations, along with a robust backtracking mechanism that verifies the feasibility of previously visited paths before replaying them, preventing unintended side effects. To further guide exploration, WebOperator generates action candidates from multiple, varied reasoning contexts to ensure diverse and robust exploration, and subsequently curates a high-quality action set by filtering out invalid actions pre-execution and merging semantically equivalent ones. Experimental results on WebArena and WebVoyager demonstrate the effectiveness of WebOperator. On WebArena, WebOperator achieves a state-of-the-art 54.6% success rate with gpt-4o, underscoring the critical advantage of integrating strategic foresight with safe execution.</p> + https://arxiv.org/abs/2512.12692 + Sun, 14 Dec 2025 13:56:54 +0000 - AI & Human Co-Improvement for Safer Co-Superintelligence - https://arxiv.org/abs/2512.05356 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.05356.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jason Weston, Jakob Foerster</p><p><b>Upvotes:</b> 8</p><p><b>Summary:</b> Self-improvement is a goal currently exciting the field of AI, but is fraught with danger, and may take time to fully achieve. We advocate that a more achievable and better goal for humanity is to maximize co-improvement: collaboration between human researchers and AIs to achieve co-superintelligence. That is, specifically targeting improving AI systems' ability to work with human researchers to conduct AI research together, from ideation to experimentation, in order to both accelerate AI research and to generally endow both AIs and humans with safer superintelligence through their symbiosis. Focusing on including human research improvement in the loop will both get us there faster, and more safely.</p> - https://arxiv.org/abs/2512.05356 - Fri, 05 Dec 2025 01:50:23 +0000 + NL2Repo-Bench: Towards Long-Horizon Repository Generation Evaluation of Coding Agents + https://arxiv.org/abs/2512.12730 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.12730.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jingzhe Ding, Shengda Long, Changxin Pu, Huan Zhou, Hongwan Gao, Xiang Gao, Chao He, Yue Hou, Fei Hu, Zhaojian Li, Weiran Shi, Zaiyuan Wang, Daoguang Zan, Chenchen Zhang, Xiaoxu Zhang, Qizhi Chen, Xianfu Cheng, Bo Deng, Qingshui Gu, Kai Hua, Juntao Lin, Pai Liu, Mingchen Li, Xuanguang Pan, Zifan Peng, Yujia Qin, Yong Shan, Zhewen Tan, Weihao Xie, Zihan Wang, Yishuo Yuan, Jiayu Zhang, Enduo Zhao, Yunfei Zhao, He Zhu, Chenyang Zou, Ming Ding, Jianpeng Jiao, Jiaheng Liu, Minghao Liu, Qian Liu, Chongyao Tao, Jian Yang, Tong Yang, Zhaoxiang Zhang, Xinjie Chen, Wenhao Huang, Ge Zhang</p><p><b>Upvotes:</b> 42</p><p><b>Summary:</b> Recent advances in coding agents suggest rapid progress toward autonomous software development, yet existing benchmarks fail to rigorously evaluate the long-horizon capabilities required to build complete software systems. Most prior evaluations focus on localized code generation, scaffolded completion, or short-term repair tasks, leaving open the question of whether agents can sustain coherent reasoning, planning, and execution over the extended horizons demanded by real-world repository construction. To address this gap, we present NL2Repo Bench, a benchmark explicitly designed to evaluate the long-horizon repository generation ability of coding agents. Given only a single natural-language requirements document and an empty workspace, agents must autonomously design the architecture, manage dependencies, implement multi-module logic, and produce a fully installable Python library. Our experiments across state-of-the-art open- and closed-source models reveal that long-horizon repository generation remains largely unsolved: even the strongest agents achieve below 40% average test pass rates and rarely complete an entire repository correctly. Detailed analysis uncovers fundamental long-horizon failure modes, including premature termination, loss of global coherence, fragile cross-file dependencies, and inadequate planning over hundreds of interaction steps. NL2Repo Bench establishes a rigorous, verifiable testbed for measuring sustained agentic competence and highlights long-horizon reasoning as a central bottleneck for the next generation of autonomous coding agents.</p> + https://arxiv.org/abs/2512.12730 + Sun, 14 Dec 2025 15:12:13 +0000 - BEAVER: An Efficient Deterministic LLM Verifier - https://arxiv.org/abs/2512.05439 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.05439.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Tarun Suresh, Nalin Wadhwa, Debangshu Banerjee, Gagandeep Singh</p><p><b>Upvotes:</b> 26</p><p><b>Summary:</b> As large language models (LLMs) transition from research prototypes to production systems, practitioners often need reliable methods to verify that model outputs satisfy required constraints. While sampling-based estimates provide an intuition of model behavior, they offer no sound guarantees. We present BEAVER, the first practical framework for computing deterministic, sound probability bounds on LLM constraint satisfaction. Given any prefix-closed semantic constraint, BEAVER systematically explores the generation space using novel token trie and frontier data structures, maintaining provably sound bounds at every iteration. We formalize the verification problem, prove soundness of our approach, and evaluate BEAVER on correctness verification, privacy verification and secure code generation tasks across multiple state of the art LLMs. BEAVER achieves 6 to 8 times tighter probability bounds and identifies 3 to 4 times more high risk instances compared to baseline methods under identical computational budgets, enabling precise characterization and risk assessment that loose bounds or empirical evaluation cannot provide.</p> - https://arxiv.org/abs/2512.05439 - Fri, 05 Dec 2025 05:34:06 +0000 + DrivePI: Spatial-aware 4D MLLM for Unified Autonomous Driving Understanding, Perception, Prediction and Planning + https://arxiv.org/abs/2512.12799 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.12799.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zhe Liu, Runhui Huang, Rui Yang, Siming Yan, Zining Wang, Lu Hou, Di Lin, Xiang Bai, Hengshuang Zhao</p><p><b>Upvotes:</b> 9</p><p><b>Summary:</b> Although multi-modal large language models (MLLMs) have shown strong capabilities across diverse domains, their application in generating fine-grained 3D perception and prediction outputs in autonomous driving remains underexplored. In this paper, we propose DrivePI, a novel spatial-aware 4D MLLM that serves as a unified Vision-Language-Action (VLA) framework that is also compatible with vision-action (VA) models. Our method jointly performs spatial understanding, 3D perception (i.e., 3D occupancy), prediction (i.e., occupancy flow), and planning (i.e., action outputs) in parallel through end-to-end optimization. To obtain both precise geometric information and rich visual appearance, our approach integrates point clouds, multi-view images, and language instructions within a unified MLLM architecture. We further develop a data engine to generate text-occupancy and text-flow QA pairs for 4D spatial understanding. Remarkably, with only a 0.5B Qwen2.5 model as MLLM backbone, DrivePI as a single unified model matches or exceeds both existing VLA models and specialized VA models. Specifically, compared to VLA models, DrivePI outperforms OpenDriveVLA-7B by 2.5% mean accuracy on nuScenes-QA and reduces collision rate by 70% over ORION (from 0.37% to 0.11%) on nuScenes. Against specialized VA models, DrivePI surpasses FB-OCC by 10.3 RayIoU for 3D occupancy on OpenOcc, reduces the mAVE from 0.591 to 0.509 for occupancy flow on OpenOcc, and achieves 32% lower L2 error than VAD (from 0.72m to 0.49m) for planning on nuScenes. Code will be available at https://github.com/happinesslz/DrivePI</p> + https://arxiv.org/abs/2512.12799 + Sun, 14 Dec 2025 18:45:54 +0000 - TED-4DGS: Temporally Activated and Embedding-based Deformation for 4DGS Compression - https://arxiv.org/abs/2512.05446 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.05446.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Cheng-Yuan Ho, He-Bi Yang, Jui-Chiu Chiang, Yu-Lun Liu, Wen-Hsiao Peng</p><p><b>Upvotes:</b> 7</p><p><b>Summary:</b> Building on the success of 3D Gaussian Splatting (3DGS) in static 3D scene representation, its extension to dynamic scenes, commonly referred to as 4DGS or dynamic 3DGS, has attracted increasing attention. However, designing more compact and efficient deformation schemes together with rate-distortion-optimized compression strategies for dynamic 3DGS representations remains an underexplored area. Prior methods either rely on space-time 4DGS with overspecified, short-lived Gaussian primitives or on canonical 3DGS with deformation that lacks explicit temporal control. To address this, we present TED-4DGS, a temporally activated and embedding-based deformation scheme for rate-distortion-optimized 4DGS compression that unifies the strengths of both families. TED-4DGS is built on a sparse anchor-based 3DGS representation. Each canonical anchor is assigned learnable temporal-activation parameters to specify its appearance and disappearance transitions over time, while a lightweight per-anchor temporal embedding queries a shared deformation bank to produce anchor-specific deformation. For rate-distortion compression, we incorporate an implicit neural representation (INR)-based hyperprior to model anchor attribute distributions, along with a channel-wise autoregressive model to capture intra-anchor correlations. With these novel elements, our scheme achieves state-of-the-art rate-distortion performance on several real-world datasets. To the best of our knowledge, this work represents one of the first attempts to pursue a rate-distortion-optimized compression framework for dynamic 3DGS representations.</p> - https://arxiv.org/abs/2512.05446 - Fri, 05 Dec 2025 05:46:35 +0000 + QwenLong-L1.5: Post-Training Recipe for Long-Context Reasoning and Memory Management + https://arxiv.org/abs/2512.12967 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.12967.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Weizhou Shen, Ziyi Yang, Chenliang Li, Zhiyuan Lu, Miao Peng, Huashan Sun, Yingcheng Shi, Shengyi Liao, Shaopeng Lai, Bo Zhang, Dayiheng Liu, Fei Huang, Jingren Zhou, Ming Yan</p><p><b>Upvotes:</b> 97</p><p><b>Summary:</b> We introduce QwenLong-L1.5, a model that achieves superior long-context reasoning capabilities through systematic post-training innovations. The key technical breakthroughs of QwenLong-L1.5 are as follows: (1) Long-Context Data Synthesis Pipeline: We develop a systematic synthesis framework that generates challenging reasoning tasks requiring multi-hop grounding over globally distributed evidence. By deconstructing documents into atomic facts and their underlying relationships, and then programmatically composing verifiable reasoning questions, our approach creates high-quality training data at scale, moving substantially beyond simple retrieval tasks to enable genuine long-range reasoning capabilities. (2) Stabilized Reinforcement Learning for Long-Context Training: To overcome the critical instability in long-context RL, we introduce task-balanced sampling with task-specific advantage estimation to mitigate reward bias, and propose Adaptive Entropy-Controlled Policy Optimization (AEPO) that dynamically regulates exploration-exploitation trade-offs. (3) Memory-Augmented Architecture for Ultra-Long Contexts: Recognizing that even extended context windows cannot accommodate arbitrarily long sequences, we develop a memory management framework with multi-stage fusion RL training that seamlessly integrates single-pass reasoning with iterative memory-based processing for tasks exceeding 4M tokens. Based on Qwen3-30B-A3B-Thinking, QwenLong-L1.5 achieves performance comparable to GPT-5 and Gemini-2.5-Pro on long-context reasoning benchmarks, surpassing its baseline by 9.90 points on average. On ultra-long tasks (1M~4M tokens), QwenLong-L1.5's memory-agent framework yields a 9.48-point gain over the agent baseline. Additionally, the acquired long-context reasoning ability translates to enhanced performance in general domains like scientific reasoning, memory tool using, and extended dialogue.</p> + https://arxiv.org/abs/2512.12967 + Mon, 15 Dec 2025 04:11:11 +0000 - ProPhy: Progressive Physical Alignment for Dynamic World Simulation - https://arxiv.org/abs/2512.05564 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.05564.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zijun Wang, Panwen Hu, Jing Wang, Terry Jingchen Zhang, Yuhao Cheng, Long Chen, Yiqiang Yan, Zutao Jiang, Hanhui Li, Xiaodan Liang</p><p><b>Upvotes:</b> 4</p><p><b>Summary:</b> Recent advances in video generation have shown remarkable potential for constructing world simulators. However, current models still struggle to produce physically consistent results, particularly when handling large-scale or complex dynamics. This limitation arises primarily because existing approaches respond isotropically to physical prompts and neglect the fine-grained alignment between generated content and localized physical cues. To address these challenges, we propose ProPhy, a Progressive Physical Alignment Framework that enables explicit physics-aware conditioning and anisotropic generation. ProPhy employs a two-stage Mixture-of-Physics-Experts (MoPE) mechanism for discriminative physical prior extraction, where Semantic Experts infer semantic-level physical principles from textual descriptions, and Refinement Experts capture token-level physical dynamics. This mechanism allows the model to learn fine-grained, physics-aware video representations that better reflect underlying physical laws. Furthermore, we introduce a physical alignment strategy that transfers the physical reasoning capabilities of vision-language models (VLMs) into the Refinement Experts, facilitating a more accurate representation of dynamic physical phenomena. Extensive experiments on physics-aware video generation benchmarks demonstrate that ProPhy produces more realistic, dynamic, and physically coherent results than existing state-of-the-art methods.</p> - https://arxiv.org/abs/2512.05564 - Fri, 05 Dec 2025 09:39:26 +0000 + Reveal Hidden Pitfalls and Navigate Next Generation of Vector Similarity Search from Task-Centric Views + https://arxiv.org/abs/2512.12980 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.12980.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Tingyang Chen, Cong Fu, Jiahua Wu, Haotian Wu, Hua Fan, Xiangyu Ke, Yunjun Gao, Yabo Ni, Anxiang Zeng</p><p><b>Upvotes:</b> 26</p><p><b>Summary:</b> Vector Similarity Search (VSS) in high-dimensional spaces is rapidly emerging as core functionality in next-generation database systems for numerous data-intensive services -- from embedding lookups in large language models (LLMs), to semantic information retrieval and recommendation engines. Current benchmarks, however, evaluate VSS primarily on the recall-latency trade-off against a ground truth defined solely by distance metrics, neglecting how retrieval quality ultimately impacts downstream tasks. This disconnect can mislead both academic research and industrial practice. We present Iceberg, a holistic benchmark suite for end-to-end evaluation of VSS methods in realistic application contexts. From a task-centric view, Iceberg uncovers the Information Loss Funnel, which identifies three principal sources of end-to-end performance degradation: (1) Embedding Loss during feature extraction; (2) Metric Misuse, where distances poorly reflect task relevance; (3) Data Distribution Sensitivity, highlighting index robustness across skews and modalities. For a more comprehensive assessment, Iceberg spans eight diverse datasets across key domains such as image classification, face recognition, text retrieval, and recommendation systems. Each dataset, ranging from 1M to 100M vectors, includes rich, task-specific labels and evaluation metrics, enabling assessment of retrieval algorithms within the full application pipeline rather than in isolation. Iceberg benchmarks 13 state-of-the-art VSS methods and re-ranks them based on application-level metrics, revealing substantial deviations from traditional rankings derived purely from recall-latency evaluations. Building on these insights, we define a set of task-centric meta-features and derive an interpretable decision tree to guide practitioners in selecting and tuning VSS methods for their specific workloads.</p> + https://arxiv.org/abs/2512.12980 + Mon, 15 Dec 2025 04:49:33 +0000 - Entropy Ratio Clipping as a Soft Global Constraint for Stable Reinforcement Learning - https://arxiv.org/abs/2512.05591 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.05591.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zhenpeng Su, Leiyu Pan, Minxuan Lv, Tiehua Mei, Zijia Lin, Yuntao Li, Wenping Hu, Ruiming Tang, Kun Gai, Guorui Zhou</p><p><b>Upvotes:</b> 16</p><p><b>Summary:</b> Large language model post-training relies on reinforcement learning to improve model capability and alignment quality. However, the off-policy training paradigm introduces distribution shift, which often pushes the policy beyond the trust region, leading to training instabilities manifested as fluctuations in policy entropy and unstable gradients. Although PPO-Clip mitigates this issue through importance clipping, it still overlooks the global distributional shift of actions. To address these challenges, we propose using the entropy ratio between the current and previous policies as a new global metric that effectively quantifies the relative change in policy exploration throughout updates. Building on this metric, we introduce an Entropy Ratio Clipping (ERC) mechanism that imposes bidirectional constraints on the entropy ratio. This stabilizes policy updates at the global distribution level and compensates for the inability of PPO-clip to regulate probability shifts of un-sampled actions. We integrate ERC into both DAPO and GPPO reinforcement learning algorithms. Experiments across multiple benchmarks show that ERC consistently improves performance.</p> - https://arxiv.org/abs/2512.05591 - Fri, 05 Dec 2025 10:26:32 +0000 + Spatial-Aware VLA Pretraining through Visual-Physical Alignment from Human Videos + https://arxiv.org/abs/2512.13080 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.13080.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yicheng Feng, Wanpeng Zhang, Ye Wang, Hao Luo, Haoqi Yuan, Sipeng Zheng, Zongqing Lu</p><p><b>Upvotes:</b> 15</p><p><b>Summary:</b> Vision-Language-Action (VLA) models provide a promising paradigm for robot learning by integrating visual perception with language-guided policy learning. However, most existing approaches rely on 2D visual inputs to perform actions in 3D physical environments, creating a significant gap between perception and action grounding. To bridge this gap, we propose a Spatial-Aware VLA Pretraining paradigm that performs explicit alignment between visual space and physical space during pretraining, enabling models to acquire 3D spatial understanding before robot policy learning. Starting from pretrained vision-language models, we leverage large-scale human demonstration videos to extract 3D visual and 3D action annotations, forming a new source of supervision that aligns 2D visual observations with 3D spatial reasoning. We instantiate this paradigm with VIPA-VLA, a dual-encoder architecture that incorporates a 3D visual encoder to augment semantic visual representations with 3D-aware features. When adapted to downstream robot tasks, VIPA-VLA achieves significantly improved grounding between 2D vision and 3D action, resulting in more robust and generalizable robotic policies.</p> + https://arxiv.org/abs/2512.13080 + Mon, 15 Dec 2025 08:31:47 +0000 - Active Video Perception: Iterative Evidence Seeking for Agentic Long Video Understanding - https://arxiv.org/abs/2512.05774 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.05774.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Ziyang Wang, Honglu Zhou, Shijie Wang, Junnan Li, Caiming Xiong, Silvio Savarese, Mohit Bansal, Michael S. Ryoo, Juan Carlos Niebles</p><p><b>Upvotes:</b> 6</p><p><b>Summary:</b> Long video understanding (LVU) is challenging because answering real-world queries often depends on sparse, temporally dispersed cues buried in hours of mostly redundant and irrelevant content. While agentic pipelines improve video reasoning capabilities, prevailing frameworks rely on a query-agnostic captioner to perceive video information, which wastes computation on irrelevant content and blurs fine-grained temporal and spatial information. Motivated by active perception theory, we argue that LVU agents should actively decide what, when, and where to observe, and continuously assess whether the current observation is sufficient to answer the query. We present Active Video Perception (AVP), an evidence-seeking framework that treats the video as an interactive environment and acquires compact, queryrelevant evidence directly from pixels. Concretely, AVP runs an iterative plan-observe-reflect process with MLLM agents. In each round, a planner proposes targeted video interactions, an observer executes them to extract time-stamped evidence, and a reflector evaluates the sufficiency of the evidence for the query, either halting with an answer or triggering further observation. Across five LVU benchmarks, AVP achieves highest performance with significant improvements. Notably, AVP outperforms the best agentic method by 5.7% in average accuracy while only requires 18.4% inference time and 12.4% input tokens.</p> - https://arxiv.org/abs/2512.05774 - Fri, 05 Dec 2025 15:03:48 +0000 + Finch: Benchmarking Finance & Accounting across Spreadsheet-Centric Enterprise Workflows + https://arxiv.org/abs/2512.13168 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.13168.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Haoyu Dong, Pengkun Zhang, Yan Gao, Xuanyu Dong, Yilin Cheng, Mingzhe Lu, Adina Yakefu, Shuxin Zheng</p><p><b>Upvotes:</b> 49</p><p><b>Summary:</b> We introduce a finance & accounting benchmark (Finch) for evaluating AI agents on real-world, enterprise-grade professional workflows -- interleaving data entry, structuring, formatting, web search, cross-file retrieval, calculation, modeling, validation, translation, visualization, and reporting. Finch is sourced from authentic enterprise workspaces at Enron (15,000 spreadsheets and 500,000 emails from 150 employees) and other financial institutions, preserving in-the-wild messiness across multimodal artifacts (text, tables, formulas, charts, code, and images) and spanning diverse domains such as budgeting, trading, and asset management. We propose a workflow construction process that combines LLM-assisted discovery with expert annotation: (1) LLM-assisted, expert-verified derivation of workflows from real-world email threads and version histories of spreadsheet files, and (2) meticulous expert annotation for workflows, requiring over 700 hours of domain-expert effort. This yields 172 composite workflows with 384 tasks, involving 1,710 spreadsheets with 27 million cells, along with PDFs and other artifacts, capturing the intrinsically messy, long-horizon, knowledge-intensive, and collaborative nature of real-world enterprise work. We conduct both human and automated evaluations of frontier AI systems including GPT 5.1, Claude Sonnet 4.5, Gemini 3 Pro, Grok 4, and Qwen 3 Max, and GPT 5.1 Pro spends 48 hours in total yet passes only 38.4% of workflows, while Claude Sonnet 4.5 passes just 25.0%. Comprehensive case studies further surface the challenges that real-world enterprise workflows pose for AI agents.</p> + https://arxiv.org/abs/2512.13168 + Mon, 15 Dec 2025 10:28:45 +0000 - SCAIL: Towards Studio-Grade Character Animation via In-Context Learning of 3D-Consistent Pose Representations - https://arxiv.org/abs/2512.05905 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.05905.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Wenhao Yan, Sheng Ye, Zhuoyi Yang, Jiayan Teng, ZhenHui Dong, Kairui Wen, Xiaotao Gu, Yong-Jin Liu, Jie Tang</p><p><b>Upvotes:</b> 19</p><p><b>Summary:</b> Achieving character animation that meets studio-grade production standards remains challenging despite recent progress. Existing approaches can transfer motion from a driving video to a reference image, but often fail to preserve structural fidelity and temporal consistency in wild scenarios involving complex motion and cross-identity animations. In this work, we present SCAIL (Studio-grade Character Animation via In-context Learning), a framework designed to address these challenges from two key innovations. First, we propose a novel 3D pose representation, providing a more robust and flexible motion signal. Second, we introduce a full-context pose injection mechanism within a diffusion-transformer architecture, enabling effective spatio-temporal reasoning over full motion sequences. To align with studio-level requirements, we develop a curated data pipeline ensuring both diversity and quality, and establish a comprehensive benchmark for systematic evaluation. Experiments show that SCAIL achieves state-of-the-art performance and advances character animation toward studio-grade reliability and realism.</p> - https://arxiv.org/abs/2512.05905 - Fri, 05 Dec 2025 17:38:55 +0000 + Toward Ambulatory Vision: Learning Visually-Grounded Active View Selection + https://arxiv.org/abs/2512.13250 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.13250.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Juil Koo, Daehyeon Choi, Sangwoo Youn, Phillip Y. Lee, Minhyuk Sung</p><p><b>Upvotes:</b> 9</p><p><b>Summary:</b> Vision Language Models (VLMs) excel at visual question answering (VQA) but remain limited to snapshot vision, reasoning from static images. In contrast, embodied agents require ambulatory vision, actively moving to obtain more informative views. We introduce Visually Grounded Active View Selection (VG-AVS), a task that selects the most informative next viewpoint using only the visual information in the current image, without relying on scene memory or external knowledge. To support this task, we construct a synthetic dataset with automatically generated paired query-target views and question-answer prompts. We also propose a framework that fine-tunes pretrained VLMs through supervised fine-tuning (SFT) followed by RL-based policy optimization. Our approach achieves strong question answering performance based on viewpoint selection and generalizes robustly to unseen synthetic and real scenes. Furthermore, incorporating our learned VG-AVS framework into existing scene-exploration-based EQA systems improves downstream question-answering accuracy.</p> + https://arxiv.org/abs/2512.13250 + Mon, 15 Dec 2025 12:04:26 +0000 - World Models That Know When They Don't Know: Controllable Video Generation with Calibrated Uncertainty - https://arxiv.org/abs/2512.05927 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.05927.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zhiting Mei, Tenny Yin, Micah Baker, Ola Shorinwa, Anirudha Majumdar</p><p><b>Upvotes:</b> 10</p><p><b>Summary:</b> Recent advances in generative video models have led to significant breakthroughs in high-fidelity video synthesis, specifically in controllable video generation where the generated video is conditioned on text and action inputs, e.g., in instruction-guided video editing and world modeling in robotics. Despite these exceptional capabilities, controllable video models often hallucinate - generating future video frames that are misaligned with physical reality - which raises serious concerns in many tasks such as robot policy evaluation and planning. However, state-of-the-art video models lack the ability to assess and express their confidence, impeding hallucination mitigation. To rigorously address this challenge, we propose C3, an uncertainty quantification (UQ) method for training continuous-scale calibrated controllable video models for dense confidence estimation at the subpatch level, precisely localizing the uncertainty in each generated video frame. Our UQ method introduces three core innovations to empower video models to estimate their uncertainty. First, our method develops a novel framework that trains video models for correctness and calibration via strictly proper scoring rules. Second, we estimate the video model's uncertainty in latent space, avoiding training instability and prohibitive training costs associated with pixel-space approaches. Third, we map the dense latent-space uncertainty to interpretable pixel-level uncertainty in the RGB space for intuitive visualization, providing high-resolution uncertainty heatmaps that identify untrustworthy regions. Through extensive experiments on large-scale robot learning datasets (Bridge and DROID) and real-world evaluations, we demonstrate that our method not only provides calibrated uncertainty estimates within the training distribution, but also enables effective out-of-distribution detection.</p> - https://arxiv.org/abs/2512.05927 - Fri, 05 Dec 2025 18:06:18 +0000 + Video Reality Test: Can AI-Generated ASMR Videos fool VLMs and Humans? + https://arxiv.org/abs/2512.13281 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.13281.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jiaqi Wang, Weijia Wu, Yi Zhan, Rui Zhao, Ming Hu, James Cheng, Wei Liu, Philip Torr, Kevin Qinghong Lin</p><p><b>Upvotes:</b> 61</p><p><b>Summary:</b> Recent advances in video generation have produced vivid content that are often indistinguishable from real videos, making AI-generated video detection an emerging societal challenge. Prior AIGC detection benchmarks mostly evaluate video without audio, target broad narrative domains, and focus on classification solely. Yet it remains unclear whether state-of-the-art video generation models can produce immersive, audio-paired videos that reliably deceive humans and VLMs. To this end, we introduce Video Reality Test, an ASMR-sourced video benchmark suite for testing perceptual realism under tight audio-visual coupling, featuring the following dimensions: (i) Immersive ASMR video-audio sources. Built on carefully curated real ASMR videos, the benchmark targets fine-grained action-object interactions with diversity across objects, actions, and backgrounds. (ii) Peer-Review evaluation. An adversarial creator-reviewer protocol where video generation models act as creators aiming to fool reviewers, while VLMs serve as reviewers seeking to identify fakeness. Our experimental findings show: The best creator Veo3.1-Fast even fools most VLMs: the strongest reviewer (Gemini 2.5-Pro) achieves only 56\% accuracy (random 50\%), far below that of human experts (81.25\%). Adding audio improves real-fake discrimination, yet superficial cues such as watermarks can still significantly mislead models. These findings delineate the current boundary of video generation realism and expose limitations of VLMs in perceptual fidelity and audio-visual consistency. Our code is available at https://github.com/video-reality-test/video-reality-test.</p> + https://arxiv.org/abs/2512.13281 + Mon, 15 Dec 2025 12:41:23 +0000 - EgoEdit: Dataset, Real-Time Streaming Model, and Benchmark for Egocentric Video Editing - https://arxiv.org/abs/2512.06065 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.06065.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Runjia Li, Moayed Haji-Ali, Ashkan Mirzaei, Chaoyang Wang, Arpit Sahni, Ivan Skorokhodov, Aliaksandr Siarohin, Tomas Jakab, Junlin Han, Sergey Tulyakov, Philip Torr, Willi Menapace</p><p><b>Upvotes:</b> 28</p><p><b>Summary:</b> We study instruction-guided editing of egocentric videos for interactive AR applications. While recent AI video editors perform well on third-person footage, egocentric views present unique challenges - including rapid egomotion and frequent hand-object interactions - that create a significant domain gap. Moreover, existing offline editing pipelines suffer from high latency, limiting real-time interaction. To address these issues, we present a complete ecosystem for egocentric video editing. First, we construct EgoEditData, a carefully designed and manually curated dataset specifically designed for egocentric editing scenarios, featuring rich hand-object interactions, while explicitly preserving hands. Second, we develop EgoEdit, an instruction-following egocentric video editor that supports real-time streaming inference on a single GPU. Finally, we introduce EgoEditBench, an evaluation suite targeting instruction faithfulness, hand and interaction preservation, and temporal stability under egomotion. Across both egocentric and general editing tasks, EgoEdit produces temporally stable, instruction-faithful results with interactive latency. It achieves clear gains on egocentric editing benchmarks-where existing methods struggle-while maintaining performance comparable to the strongest baselines on general editing tasks. EgoEditData and EgoEditBench will be made public for the research community. See our website at https://snap-research.github.io/EgoEdit</p> - https://arxiv.org/abs/2512.06065 - Fri, 05 Dec 2025 18:57:05 +0000 + ShowTable: Unlocking Creative Table Visualization with Collaborative Reflection and Refinement + https://arxiv.org/abs/2512.13303 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.13303.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zhihang Liu, Xiaoyi Bao, Pandeng Li, Junjie Zhou, Zhaohe Liao, Yefei He, Kaixun Jiang, Chen-Wei Xie, Yun Zheng, Hongtao Xie</p><p><b>Upvotes:</b> 16</p><p><b>Summary:</b> While existing generation and unified models excel at general image generation, they struggle with tasks requiring deep reasoning, planning, and precise data-to-visual mapping abilities beyond general scenarios. To push beyond the existing limitations, we introduce a new and challenging task: creative table visualization, requiring the model to generate an infographic that faithfully and aesthetically visualizes the data from a given table. To address this challenge, we propose ShowTable, a pipeline that synergizes MLLMs with diffusion models via a progressive self-correcting process. The MLLM acts as the central orchestrator for reasoning the visual plan and judging visual errors to provide refined instructions, the diffusion execute the commands from MLLM, achieving high-fidelity results. To support this task and our pipeline, we introduce three automated data construction pipelines for training different modules. Furthermore, we introduce TableVisBench, a new benchmark with 800 challenging instances across 5 evaluation dimensions, to assess performance on this task. Experiments demonstrate that our pipeline, instantiated with different models, significantly outperforms baselines, highlighting its effective multi-modal reasoning, generation, and error correction capabilities.</p> + https://arxiv.org/abs/2512.13303 + Mon, 15 Dec 2025 13:21:50 +0000 - EditThinker: Unlocking Iterative Reasoning for Any Image Editor - https://arxiv.org/abs/2512.05965 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.05965.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Hongyu Li, Manyuan Zhang, Dian Zheng, Ziyu Guo, Yimeng Jia, Kaituo Feng, Hao Yu, Yexin Liu, Yan Feng, Peng Pei, Xunliang Cai, Linjiang Huang, Hongsheng Li, Si Liu</p><p><b>Upvotes:</b> 36</p><p><b>Summary:</b> Instruction-based image editing has emerged as a prominent research area, which, benefiting from image generation foundation models, have achieved high aesthetic quality, making instruction-following capability the primary challenge. Existing approaches improve instruction adherence via supervised or reinforcement learning, yet single-turn success rates remain limited due to inherent stochasticity and a lack of deliberation. In this work, we propose a deliberative editing framework to 'think' while they edit, which simulates the human cognitive loop by iteratively executing a Think-while-Edit cycle: Critiquing results and Refining instructions , followed by Repeating the generation until satisfactory. Specifically, we train a single MLLM, EditThinker, to act as the reasoning engine of this framework, which jointly produce the critique score, reasoning process, and refined instructions. We employ reinforcement learning to align the EditThinker's thinking with its editing, thereby generating more targeted instruction improvements. Extensive experiments on four benchmarks demonstrate that our approach significantly improves the instruction-following capability of any image editing model by a large margin. We will release our data construction framework, datasets, and models to benefit the community.</p> - https://arxiv.org/abs/2512.05965 - Fri, 05 Dec 2025 18:58:09 +0000 + KlingAvatar 2.0 Technical Report + https://arxiv.org/abs/2512.13313 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.13313.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Kling Team, Jialu Chen, Yikang Ding, Zhixue Fang, Kun Gai, Yuan Gao, Kang He, Jingyun Hua, Boyuan Jiang, Mingming Lao, Xiaohan Li, Hui Liu, Jiwen Liu, Xiaoqiang Liu, Yuan Liu, Shun Lu, Yongsen Mao, Yingchao Shao, Huafeng Shi, Xiaoyu Shi, Peiqin Sun, Songlin Tang, Pengfei Wan, Chao Wang, Xuebo Wang, Haoxian Zhang, Yuanxing Zhang, Yan Zhou</p><p><b>Upvotes:</b> 40</p><p><b>Summary:</b> Avatar video generation models have achieved remarkable progress in recent years. However, prior work exhibits limited efficiency in generating long-duration high-resolution videos, suffering from temporal drifting, quality degradation, and weak prompt following as video length increases. To address these challenges, we propose KlingAvatar 2.0, a spatio-temporal cascade framework that performs upscaling in both spatial resolution and temporal dimension. The framework first generates low-resolution blueprint video keyframes that capture global semantics and motion, and then refines them into high-resolution, temporally coherent sub-clips using a first-last frame strategy, while retaining smooth temporal transitions in long-form videos. To enhance cross-modal instruction fusion and alignment in extended videos, we introduce a Co-Reasoning Director composed of three modality-specific large language model (LLM) experts. These experts reason about modality priorities and infer underlying user intent, converting inputs into detailed storylines through multi-turn dialogue. A Negative Director further refines negative prompts to improve instruction alignment. Building on these components, we extend the framework to support ID-specific multi-character control. Extensive experiments demonstrate that our model effectively addresses the challenges of efficient, multimodally aligned long-form high-resolution video generation, delivering enhanced visual clarity, realistic lip-teeth rendering with accurate lip synchronization, strong identity preservation, and coherent multimodal instruction following.</p> + https://arxiv.org/abs/2512.13313 + Mon, 15 Dec 2025 13:30:51 +0000 - Rethinking Training Dynamics in Scale-wise Autoregressive Generation - https://arxiv.org/abs/2512.06421 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.06421.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Gengze Zhou, Chongjian Ge, Hao Tan, Feng Liu, Yicong Hong</p><p><b>Upvotes:</b> 4</p><p><b>Summary:</b> Recent advances in autoregressive (AR) generative models have produced increasingly powerful systems for media synthesis. Among them, next-scale prediction has emerged as a popular paradigm, where models generate images in a coarse-to-fine manner. However, scale-wise AR models suffer from exposure bias, which undermines generation quality. We identify two primary causes of this issue: (1) train-test mismatch, where the model must rely on its own imperfect predictions during inference, and (2) imbalance in scale-wise learning difficulty, where certain scales exhibit disproportionately higher optimization complexity. Through a comprehensive analysis of training dynamics, we propose Self-Autoregressive Refinement (SAR) to address these limitations. SAR introduces a Stagger-Scale Rollout (SSR) mechanism that performs lightweight autoregressive rollouts to expose the model to its own intermediate predictions, thereby aligning train-test patterns, and a complementary Contrastive Student-Forcing Loss (CSFL) that provides adequate supervision for self-generated contexts to ensure stable training. Experimental results show that applying SAR to pretrained AR models consistently improves generation quality with minimal computational overhead. For instance, SAR yields a 5.2% FID reduction on FlexVAR-d16 trained on ImageNet 256 within 10 epochs (5 hours on 32xA100 GPUs). Given its efficiency, scalability, and effectiveness, we expect SAR to serve as a reliable post-training method for visual autoregressive generation.</p> - https://arxiv.org/abs/2512.06421 - Sat, 06 Dec 2025 12:41:42 +0000 + Differentiable Evolutionary Reinforcement Learning + https://arxiv.org/abs/2512.13399 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.13399.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Sitao Cheng, Tianle Li, Xuhan Huang, Xunjian Yin, Difan Zou</p><p><b>Upvotes:</b> 18</p><p><b>Summary:</b> The design of effective reward functions presents a central and often arduous challenge in reinforcement learning (RL), particularly when developing autonomous agents for complex reasoning tasks. While automated reward optimization approaches exist, they typically rely on derivative-free evolutionary heuristics that treat the reward function as a black box, failing to capture the causal relationship between reward structure and task performance. To bridge this gap, we propose Differentiable Evolutionary Reinforcement Learning (DERL), a bilevel framework that enables the autonomous discovery of optimal reward signals. In DERL, a Meta-Optimizer evolves a reward function (i.e., Meta-Reward) by composing structured atomic primitives, guiding the training of an inner-loop policy. Crucially, unlike previous evolution, DERL is differentiable in its metaoptimization: it treats the inner-loop validation performance as a signal to update the Meta-Optimizer via reinforcement learning. This allows DERL to approximate the "meta-gradient" of task success, progressively learning to generate denser and more actionable feedback. We validate DERL across three distinct domains: robotic agent (ALFWorld), scientific simulation (ScienceWorld), and mathematical reasoning (GSM8k, MATH). Experimental results show that DERL achieves state-of-the-art performance on ALFWorld and ScienceWorld, significantly outperforming methods relying on heuristic rewards, especially in out-of-distribution scenarios. Analysis of the evolutionary trajectory demonstrates that DERL successfully captures the intrinsic structure of tasks, enabling selfimproving agent alignment without human intervention.</p> + https://arxiv.org/abs/2512.13399 + Mon, 15 Dec 2025 14:50:08 +0000 - Embodied Referring Expression Comprehension in Human-Robot Interaction - https://arxiv.org/abs/2512.06558 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.06558.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Md Mofijul Islam, Alexi Gladstone, Sujan Sarker, Ganesh Nanduru, Md Fahim, Keyan Du, Aman Chadha, Tariq Iqbal</p><p><b>Upvotes:</b> 3</p><p><b>Summary:</b> As robots enter human workspaces, there is a crucial need for them to comprehend embodied human instructions, enabling intuitive and fluent human-robot interaction (HRI). However, accurate comprehension is challenging due to a lack of large-scale datasets that capture natural embodied interactions in diverse HRI settings. Existing datasets suffer from perspective bias, single-view collection, inadequate coverage of nonverbal gestures, and a predominant focus on indoor environments. To address these issues, we present the Refer360 dataset, a large-scale dataset of embodied verbal and nonverbal interactions collected across diverse viewpoints in both indoor and outdoor settings. Additionally, we introduce MuRes, a multimodal guided residual module designed to improve embodied referring expression comprehension. MuRes acts as an information bottleneck, extracting salient modality-specific signals and reinforcing them into pre-trained representations to form complementary features for downstream tasks. We conduct extensive experiments on four HRI datasets, including the Refer360 dataset, and demonstrate that current multimodal models fail to capture embodied interactions comprehensively; however, augmenting them with MuRes consistently improves performance. These findings establish Refer360 as a valuable benchmark and exhibit the potential of guided residual learning to advance embodied referring expression comprehension in robots operating within human environments.</p> - https://arxiv.org/abs/2512.06558 - Sat, 06 Dec 2025 20:10:06 +0000 + Seedance 1.5 pro: A Native Audio-Visual Joint Generation Foundation Model + https://arxiv.org/abs/2512.13507 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.13507.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Heyi Chen, Siyan Chen, Xin Chen, Yanfei Chen, Ying Chen, Zhuo Chen, Feng Cheng, Tianheng Cheng, Xinqi Cheng, Xuyan Chi, Jian Cong, Jing Cui, Qinpeng Cui, Qide Dong, Junliang Fan, Jing Fang, Zetao Fang, Chengjian Feng, Han Feng, Mingyuan Gao, Yu Gao, Dong Guo, Qiushan Guo, Boyang Hao, Qingkai Hao, Bibo He, Qian He, Tuyen Hoang, Ruoqing Hu, Xi Hu, Weilin Huang, Zhaoyang Huang, Zhongyi Huang, Donglei Ji, Siqi Jiang, Wei Jiang, Yunpu Jiang, Zhuo Jiang, Ashley Kim, Jianan Kong, Zhichao Lai, Shanshan Lao, Yichong Leng, Ai Li, Feiya Li, Gen Li, Huixia Li, JiaShi Li, Liang Li, Ming Li, Shanshan Li, Tao Li, Xian Li, Xiaojie Li, Xiaoyang Li, Xingxing Li, Yameng Li, Yifu Li, Yiying Li, Chao Liang, Han Liang, Jianzhong Liang, Ying Liang, Zhiqiang Liang, Wang Liao, Yalin Liao, Heng Lin, Kengyu Lin, Shanchuan Lin, Xi Lin, Zhijie Lin, Feng Ling, Fangfang Liu, Gaohong Liu, Jiawei Liu, Jie Liu, Jihao Liu, Shouda Liu, Shu Liu, Sichao Liu, Songwei Liu, Xin Liu, Xue Liu, Yibo Liu, Zikun Liu, Zuxi Liu, Junlin Lyu, Lecheng Lyu, Qian Lyu, Han Mu, Xiaonan Nie, Jingzhe Ning, Xitong Pan, Yanghua Peng, Lianke Qin, Xueqiong Qu, Yuxi Ren, Kai Shen, Guang Shi, Lei Shi, Yan Song, Yinglong Song, Fan Sun, Li Sun, Renfei Sun, Yan Sun, Zeyu Sun, Wenjing Tang, Yaxue Tang, Zirui Tao, Feng Wang, Furui Wang, Jinran Wang, Junkai Wang, Ke Wang, Kexin Wang, Qingyi Wang, Rui Wang, Sen Wang, Shuai Wang, Tingru Wang, Weichen Wang, Xin Wang, Yanhui Wang, Yue Wang, Yuping Wang, Yuxuan Wang, Ziyu Wang, Guoqiang Wei, Wanru Wei, Di Wu, Guohong Wu, Hanjie Wu, Jian Wu, Jie Wu, Ruolan Wu, Xinglong Wu, Yonghui Wu, Ruiqi Xia, Liang Xiang, Fei Xiao, XueFeng Xiao, Pan Xie, Shuangyi Xie, Shuang Xu, Jinlan Xue, Shen Yan, Bangbang Yang, Ceyuan Yang, Jiaqi Yang, Runkai Yang, Tao Yang, Yang Yang, Yihang Yang, ZhiXian Yang, Ziyan Yang, Songting Yao, Yifan Yao, Zilyu Ye, Bowen Yu, Jian Yu, Chujie Yuan, Linxiao Yuan, Sichun Zeng, Weihong Zeng, Xuejiao Zeng, Yan Zeng, Chuntao Zhang, Heng Zhang, Jingjie Zhang, Kuo Zhang, Liang Zhang, Liying Zhang, Manlin Zhang, Ting Zhang, Weida Zhang, Xiaohe Zhang, Xinyan Zhang, Yan Zhang, Yuan Zhang, Zixiang Zhang, Fengxuan Zhao, Huating Zhao, Yang Zhao, Hao Zheng, Jianbin Zheng, Xiaozheng Zheng, Yangyang Zheng, Yijie Zheng, Jiexin Zhou, Jiahui Zhu, Kuan Zhu, Shenhan Zhu, Wenjia Zhu, Benhui Zou, Feilong Zuo</p><p><b>Upvotes:</b> 33</p><p><b>Summary:</b> Recent strides in video generation have paved the way for unified audio-visual generation. In this work, we present Seedance 1.5 pro, a foundational model engineered specifically for native, joint audio-video generation. Leveraging a dual-branch Diffusion Transformer architecture, the model integrates a cross-modal joint module with a specialized multi-stage data pipeline, achieving exceptional audio-visual synchronization and superior generation quality. To ensure practical utility, we implement meticulous post-training optimizations, including Supervised Fine-Tuning (SFT) on high-quality datasets and Reinforcement Learning from Human Feedback (RLHF) with multi-dimensional reward models. Furthermore, we introduce an acceleration framework that boosts inference speed by over 10X. Seedance 1.5 pro distinguishes itself through precise multilingual and dialect lip-syncing, dynamic cinematic camera control, and enhanced narrative coherence, positioning it as a robust engine for professional-grade content creation. Seedance 1.5 pro is now accessible on Volcano Engine at https://console.volcengine.com/ark/region:ark+cn-beijing/experience/vision?type=GenVideo.</p> + https://arxiv.org/abs/2512.13507 + Mon, 15 Dec 2025 16:36:52 +0000 - OmniSafeBench-MM: A Unified Benchmark and Toolbox for Multimodal Jailbreak Attack-Defense Evaluation - https://arxiv.org/abs/2512.06589 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.06589.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Xiaojun Jia, Jie Liao, Qi Guo, Teng Ma, Simeng Qin, Ranjie Duan, Tianlin Li, Yihao Huang, Zhitao Zeng, Dongxian Wu, Yiming Li, Wenqi Ren, Xiaochun Cao, Yang Liu</p><p><b>Upvotes:</b> 16</p><p><b>Summary:</b> Recent advances in multi-modal large language models (MLLMs) have enabled unified perception-reasoning capabilities, yet these systems remain highly vulnerable to jailbreak attacks that bypass safety alignment and induce harmful behaviors. Existing benchmarks such as JailBreakV-28K, MM-SafetyBench, and HADES provide valuable insights into multi-modal vulnerabilities, but they typically focus on limited attack scenarios, lack standardized defense evaluation, and offer no unified, reproducible toolbox. To address these gaps, we introduce OmniSafeBench-MM, which is a comprehensive toolbox for multi-modal jailbreak attack-defense evaluation. OmniSafeBench-MM integrates 13 representative attack methods, 15 defense strategies, and a diverse dataset spanning 9 major risk domains and 50 fine-grained categories, structured across consultative, imperative, and declarative inquiry types to reflect realistic user intentions. Beyond data coverage, it establishes a three-dimensional evaluation protocol measuring (1) harmfulness, distinguished by a granular, multi-level scale ranging from low-impact individual harm to catastrophic societal threats, (2) intent alignment between responses and queries, and (3) response detail level, enabling nuanced safety-utility analysis. We conduct extensive experiments on 10 open-source and 8 closed-source MLLMs to reveal their vulnerability to multi-modal jailbreak. By unifying data, methodology, and evaluation into an open-source, reproducible platform, OmniSafeBench-MM provides a standardized foundation for future research. The code is released at https://github.com/jiaxiaojunQAQ/OmniSafeBench-MM.</p> - https://arxiv.org/abs/2512.06589 - Sat, 06 Dec 2025 22:56:29 +0000 + Memory in the Age of AI Agents + https://arxiv.org/abs/2512.13564 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.13564.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yuyang Hu, Shichun Liu, Yanwei Yue, Guibin Zhang, Boyang Liu, Fangyi Zhu, Jiahang Lin, Honglin Guo, Shihan Dou, Zhiheng Xi, Senjie Jin, Jiejun Tan, Yanbin Yin, Jiongnan Liu, Zeyu Zhang, Zhongxiang Sun, Yutao Zhu, Hao Sun, Boci Peng, Zhenrong Cheng, Xuanbo Fan, Jiaxin Guo, Xinlei Yu, Zhenhong Zhou, Zewen Hu, Jiahao Huo, Junhao Wang, Yuwei Niu, Yu Wang, Zhenfei Yin, Xiaobin Hu, Yue Liao, Qiankun Li, Kun Wang, Wangchunshu Zhou, Yixin Liu, Dawei Cheng, Qi Zhang, Tao Gui, Shirui Pan, Yan Zhang, Philip Torr, Zhicheng Dou, Ji-Rong Wen, Xuanjing Huang, Yu-Gang Jiang, Shuicheng Yan</p><p><b>Upvotes:</b> 101</p><p><b>Summary:</b> Memory has emerged, and will continue to remain, a core capability of foundation model-based agents. As research on agent memory rapidly expands and attracts unprecedented attention, the field has also become increasingly fragmented. Existing works that fall under the umbrella of agent memory often differ substantially in their motivations, implementations, and evaluation protocols, while the proliferation of loosely defined memory terminologies has further obscured conceptual clarity. Traditional taxonomies such as long/short-term memory have proven insufficient to capture the diversity of contemporary agent memory systems. This work aims to provide an up-to-date landscape of current agent memory research. We begin by clearly delineating the scope of agent memory and distinguishing it from related concepts such as LLM memory, retrieval augmented generation (RAG), and context engineering. We then examine agent memory through the unified lenses of forms, functions, and dynamics. From the perspective of forms, we identify three dominant realizations of agent memory, namely token-level, parametric, and latent memory. From the perspective of functions, we propose a finer-grained taxonomy that distinguishes factual, experiential, and working memory. From the perspective of dynamics, we analyze how memory is formed, evolved, and retrieved over time. To support practical development, we compile a comprehensive summary of memory benchmarks and open-source frameworks. Beyond consolidation, we articulate a forward-looking perspective on emerging research frontiers, including memory automation, reinforcement learning integration, multimodal memory, multi-agent memory, and trustworthiness issues. We hope this survey serves not only as a reference for existing work, but also as a conceptual foundation for rethinking memory as a first-class primitive in the design of future agentic intelligence.</p> + https://arxiv.org/abs/2512.13564 + Mon, 15 Dec 2025 17:22:34 +0000 - MIND-V: Hierarchical Video Generation for Long-Horizon Robotic Manipulation with RL-based Physical Alignment - https://arxiv.org/abs/2512.06628 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.06628.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Ruicheng Zhang, Mingyang Zhang, Jun Zhou, Zhangrui Guo, Xiaofan Liu, Zunnan Xu, Zhizhou Zhong, Puxin Yan, Haocheng Luo, Xiu Li</p><p><b>Upvotes:</b> 12</p><p><b>Summary:</b> Embodied imitation learning is constrained by the scarcity of diverse, long-horizon robotic manipulation data. Existing video generation models for this domain are limited to synthesizing short clips of simple actions and often rely on manually defined trajectories. To this end, we introduce MIND-V, a hierarchical framework designed to synthesize physically plausible and logically coherent videos of long-horizon robotic manipulation. Inspired by cognitive science, MIND-V bridges high-level reasoning with pixel-level synthesis through three core components: a Semantic Reasoning Hub (SRH) that leverages a pre-trained vision-language model for task planning; a Behavioral Semantic Bridge (BSB) that translates abstract instructions into domain-invariant representations; and a Motor Video Generator (MVG) for conditional video rendering. MIND-V employs Staged Visual Future Rollouts, a test-time optimization strategy to enhance long-horizon robustness. To align the generated videos with physical laws, we introduce a GRPO reinforcement learning post-training phase guided by a novel Physical Foresight Coherence (PFC) reward. PFC leverages the V-JEPA world model to enforce physical plausibility by aligning the predicted and actual dynamic evolutions in the feature space. MIND-V demonstrates state-of-the-art performance in long-horizon robotic manipulation video generation, establishing a scalable and controllable paradigm for embodied data synthesis.</p> - https://arxiv.org/abs/2512.06628 - Sun, 07 Dec 2025 02:28:06 +0000 + ReFusion: A Diffusion Large Language Model with Parallel Autoregressive Decoding + https://arxiv.org/abs/2512.13586 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.13586.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jia-Nan Li, Jian Guan, Wei Wu, Chongxuan Li</p><p><b>Upvotes:</b> 87</p><p><b>Summary:</b> Autoregressive models (ARMs) are hindered by slow sequential inference. While masked diffusion models (MDMs) offer a parallel alternative, they suffer from critical drawbacks: high computational overhead from precluding Key-Value (KV) caching, and incoherent generation arising from learning dependencies over an intractable space of token combinations. To address these limitations, we introduce ReFusion, a novel masked diffusion model that achieves superior performance and efficiency by elevating parallel decoding from the token level to a higher slot level, where each slot is a fixed-length, contiguous sub-sequence. This is achieved through an iterative ``plan-and-infill'' decoding process: a diffusion-based planning step first identifies a set of weakly dependent slots, and an autoregressive infilling step then decodes these selected slots in parallel. The slot-based design simultaneously unlocks full KV cache reuse with a unified causal framework and reduces the learning complexity from the token combination space to a manageable slot-level permutation space. Extensive experiments on seven diverse benchmarks show that ReFusion not only overwhelmingly surpasses prior MDMs with 34% performance gains and an over 18times speedup on average, but also bridges the performance gap to strong ARMs while maintaining a 2.33times average speedup.</p> + https://arxiv.org/abs/2512.13586 + Mon, 15 Dec 2025 17:41:19 +0000 - DoVer: Intervention-Driven Auto Debugging for LLM Multi-Agent Systems - https://arxiv.org/abs/2512.06749 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.06749.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Ming Ma, Jue Zhang, Fangkai Yang, Yu Kang, Qingwei Lin, Saravan Rajmohan, Dongmei Zhang</p><p><b>Upvotes:</b> 25</p><p><b>Summary:</b> Large language model (LLM)-based multi-agent systems are challenging to debug because failures often arise from long, branching interaction traces. The prevailing practice is to leverage LLMs for log-based failure localization, attributing errors to a specific agent and step. However, this paradigm has two key limitations: (i) log-only debugging lacks validation, producing untested hypotheses, and (ii) single-step or single-agent attribution is often ill-posed, as we find that multiple distinct interventions can independently repair the failed task. To address the first limitation, we introduce DoVer, an intervention-driven debugging framework, which augments hypothesis generation with active verification through targeted interventions (e.g., editing messages, altering plans). For the second limitation, rather than evaluating on attribution accuracy, we focus on measuring whether the system resolves the failure or makes quantifiable progress toward task success, reflecting a more outcome-oriented view of debugging. Within the Magnetic-One agent framework, on the datasets derived from GAIA and AssistantBench, DoVer flips 18-28% of failed trials into successes, achieves up to 16% milestone progress, and validates or refutes 30-60% of failure hypotheses. DoVer also performs effectively on a different dataset (GSMPlus) and agent framework (AG2), where it recovers 49% of failed trials. These results highlight intervention as a practical mechanism for improving reliability in agentic systems and open opportunities for more robust, scalable debugging methods for LLM-based multi-agent systems. Project website and code will be available at https://aka.ms/DoVer.</p> - https://arxiv.org/abs/2512.06749 - Sun, 07 Dec 2025 09:23:48 +0000 + Image Diffusion Preview with Consistency Solver + https://arxiv.org/abs/2512.13592 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.13592.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Fu-Yun Wang, Hao Zhou, Liangzhe Yuan, Sanghyun Woo, Boqing Gong, Bohyung Han, Ming-Hsuan Yang, Han Zhang, Yukun Zhu, Ting Liu, Long Zhao</p><p><b>Upvotes:</b> 6</p><p><b>Summary:</b> The slow inference process of image diffusion models significantly degrades interactive user experiences. To address this, we introduce Diffusion Preview, a novel paradigm employing rapid, low-step sampling to generate preliminary outputs for user evaluation, deferring full-step refinement until the preview is deemed satisfactory. Existing acceleration methods, including training-free solvers and post-training distillation, struggle to deliver high-quality previews or ensure consistency between previews and final outputs. We propose ConsistencySolver derived from general linear multistep methods, a lightweight, trainable high-order solver optimized via Reinforcement Learning, that enhances preview quality and consistency. Experimental results demonstrate that ConsistencySolver significantly improves generation quality and consistency in low-step scenarios, making it ideal for efficient preview-and-refine workflows. Notably, it achieves FID scores on-par with Multistep DPM-Solver using 47% fewer steps, while outperforming distillation baselines. Furthermore, user studies indicate our approach reduces overall user interaction time by nearly 50% while maintaining generation quality. Code is available at https://github.com/G-U-N/consolver.</p> + https://arxiv.org/abs/2512.13592 + Mon, 15 Dec 2025 17:47:49 +0000 - From Next-Token to Next-Block: A Principled Adaptation Path for Diffusion LLMs - https://arxiv.org/abs/2512.06776 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.06776.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yuchuan Tian, Yuchen Liang, Jiacheng Sun, Shuo Zhang, Guangwen Yang, Yingte Shu, Sibo Fang, Tianyu Guo, Kai Han, Chao Xu, Hanting Chen, Xinghao Chen, Yunhe Wang</p><p><b>Upvotes:</b> 21</p><p><b>Summary:</b> Large language models (LLMs) excel at generation but dominant autoregressive (AR) decoding is inherently sequential, creating a throughput bottleneck. Diffusion Language Models (DLMs)--especially block-wise variants--enable parallel generation and intra-block bidirectional reasoning, yet training large DLMs from scratch is costly and wastes the knowledge in mature AR checkpoints. Prior "adaptation" attempts either modify logits or randomly grow attention masks to full-sequence diffusion, or simply transplant AR weights into a block-diffusion recipe, leaving a fundamental mismatch between AR causality and block-wise bidirectionality unaddressed. We reframe adaptation as a intra-paradigm path from AR to Block-Diffusion by viewing AR as Block-Diffusion with blocksize=1. Concretely, we design the pathway of adaptation as follows: we use a context-causal attention mask (causal in context, bidirectional only within the active block), an efficient parallel adaptation procedure, an auxiliary AR loss to maximize data utilization and retain pretrained knowledge, and gradual increment of the generation block size. The recipe integrates cleanly with masked block-diffusion and maintains train-inference consistency. Built on these components, NBDiff-7B (Base and Instruct) could inherit the long-context modeling and reasoning capabilities, and achieve state-of-the-art performance among the 7B-class DLMs, delivering strong gains on general-knowledge, math, and code benchmarks over strong baselines. These results demonstrate that principled AR-to-block-diffusion adaptation is an effective and compute-efficient alternative to training DLMs from scratch. Codes: https://github.com/YuchuanTian/NBDiff.</p> - https://arxiv.org/abs/2512.06776 - Sun, 07 Dec 2025 10:28:21 +0000 + LongVie 2: Multimodal Controllable Ultra-Long Video World Model + https://arxiv.org/abs/2512.13604 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.13604.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jianxiong Gao, Zhaoxi Chen, Xian Liu, Junhao Zhuang, Chengming Xu, Jianfeng Feng, Yu Qiao, Yanwei Fu, Chenyang Si, Ziwei Liu</p><p><b>Upvotes:</b> 68</p><p><b>Summary:</b> Building video world models upon pretrained video generation systems represents an important yet challenging step toward general spatiotemporal intelligence. A world model should possess three essential properties: controllability, long-term visual quality, and temporal consistency. To this end, we take a progressive approach-first enhancing controllability and then extending toward long-term, high-quality generation. We present LongVie 2, an end-to-end autoregressive framework trained in three stages: (1) Multi-modal guidance, which integrates dense and sparse control signals to provide implicit world-level supervision and improve controllability; (2) Degradation-aware training on the input frame, bridging the gap between training and long-term inference to maintain high visual quality; and (3) History-context guidance, which aligns contextual information across adjacent clips to ensure temporal consistency. We further introduce LongVGenBench, a comprehensive benchmark comprising 100 high-resolution one-minute videos covering diverse real-world and synthetic environments. Extensive experiments demonstrate that LongVie 2 achieves state-of-the-art performance in long-range controllability, temporal coherence, and visual fidelity, and supports continuous video generation lasting up to five minutes, marking a significant step toward unified video world modeling.</p> + https://arxiv.org/abs/2512.13604 + Mon, 15 Dec 2025 17:59:58 +0000 - Boosting Unsupervised Video Instance Segmentation with Automatic Quality-Guided Self-Training - https://arxiv.org/abs/2512.06864 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.06864.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Kaixuan Lu, Mehmet Onurcan Kaya, Dim P. Papadopoulos</p><p><b>Upvotes:</b> 12</p><p><b>Summary:</b> Video Instance Segmentation (VIS) faces significant annotation challenges due to its dual requirements of pixel-level masks and temporal consistency labels. While recent unsupervised methods like VideoCutLER eliminate optical flow dependencies through synthetic data, they remain constrained by the synthetic-to-real domain gap. We present AutoQ-VIS, a novel unsupervised framework that bridges this gap through quality-guided self-training. Our approach establishes a closed-loop system between pseudo-label generation and automatic quality assessment, enabling progressive adaptation from synthetic to real videos. Experiments demonstrate state-of-the-art performance with 52.6 AP_{50} on YouTubeVIS-2019 val set, surpassing the previous state-of-the-art VideoCutLER by 4.4%, while requiring no human annotations. This demonstrates the viability of quality-aware self-training for unsupervised VIS. We will release the code at https://github.com/wcbup/AutoQ-VIS.</p> - https://arxiv.org/abs/2512.06864 - Sun, 07 Dec 2025 14:37:12 +0000 + Nemotron-Cascade: Scaling Cascaded Reinforcement Learning for General-Purpose Reasoning Models + https://arxiv.org/abs/2512.13607 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.13607.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Boxin Wang, Chankyu Lee, Nayeon Lee, Sheng-Chieh Lin, Wenliang Dai, Yang Chen, Yangyi Chen, Zhuolin Yang, Zihan Liu, Mohammad Shoeybi, Bryan Catanzaro, Wei Ping</p><p><b>Upvotes:</b> 24</p><p><b>Summary:</b> Building general-purpose reasoning models with reinforcement learning (RL) entails substantial cross-domain heterogeneity, including large variation in inference-time response lengths and verification latency. Such variability complicates the RL infrastructure, slows training, and makes training curriculum (e.g., response length extension) and hyperparameter selection challenging. In this work, we propose cascaded domain-wise reinforcement learning (Cascade RL) to develop general-purpose reasoning models, Nemotron-Cascade, capable of operating in both instruct and deep thinking modes. Departing from conventional approaches that blend heterogeneous prompts from different domains, Cascade RL orchestrates sequential, domain-wise RL, reducing engineering complexity and delivering state-of-the-art performance across a wide range of benchmarks. Notably, RLHF for alignment, when used as a pre-step, boosts the model's reasoning ability far beyond mere preference optimization, and subsequent domain-wise RLVR stages rarely degrade the benchmark performance attained in earlier domains and may even improve it (see an illustration in Figure 1). Our 14B model, after RL, outperforms its SFT teacher, DeepSeek-R1-0528, on LiveCodeBench v5/v6/Pro and achieves silver-medal performance in the 2025 International Olympiad in Informatics (IOI). We transparently share our training and data recipes.</p> + https://arxiv.org/abs/2512.13607 + Mon, 15 Dec 2025 18:02:35 +0000 - Scaling Zero-Shot Reference-to-Video Generation - https://arxiv.org/abs/2512.06905 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.06905.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zijian Zhou, Shikun Liu, Haozhe Liu, Haonan Qiu, Zhaochong An, Weiming Ren, Zhiheng Liu, Xiaoke Huang, Kam Woh Ng, Tian Xie, Xiao Han, Yuren Cong, Hang Li, Chuyan Zhu, Aditya Patel, Tao Xiang, Sen He</p><p><b>Upvotes:</b> 28</p><p><b>Summary:</b> Reference-to-video (R2V) generation aims to synthesize videos that align with a text prompt while preserving the subject identity from reference images. However, current R2V methods are hindered by the reliance on explicit reference image-video-text triplets, whose construction is highly expensive and difficult to scale. We bypass this bottleneck by introducing Saber, a scalable zero-shot framework that requires no explicit R2V data. Trained exclusively on video-text pairs, Saber employs a masked training strategy and a tailored attention-based model design to learn identity-consistent and reference-aware representations. Mask augmentation techniques are further integrated to mitigate copy-paste artifacts common in reference-to-video generation. Moreover, Saber demonstrates remarkable generalization capabilities across a varying number of references and achieves superior performance on the OpenS2V-Eval benchmark compared to methods trained with R2V data.</p> - https://arxiv.org/abs/2512.06905 - Sun, 07 Dec 2025 16:10:25 +0000 + RoboTracer: Mastering Spatial Trace with Reasoning in Vision-Language Models for Robotics + https://arxiv.org/abs/2512.13660 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.13660.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Enshen Zhou, Cheng Chi, Yibo Li, Jingkun An, Jiayuan Zhang, Shanyu Rong, Yi Han, Yuheng Ji, Mengzhen Liu, Pengwei Wang, Zhongyuan Wang, Lu Sheng, Shanghang Zhang</p><p><b>Upvotes:</b> 36</p><p><b>Summary:</b> Spatial tracing, as a fundamental embodied interaction ability for robots, is inherently challenging as it requires multi-step metric-grounded reasoning compounded with complex spatial referring and real-world metric measurement. However, existing methods struggle with this compositional task. To this end, we propose RoboTracer, a 3D-aware VLM that first achieves both 3D spatial referring and measuring via a universal spatial encoder and a regression-supervised decoder to enhance scale awareness during supervised fine-tuning (SFT). Moreover, RoboTracer advances multi-step metric-grounded reasoning via reinforcement fine-tuning (RFT) with metric-sensitive process rewards, supervising key intermediate perceptual cues to accurately generate spatial traces. To support SFT and RFT training, we introduce TraceSpatial, a large-scale dataset of 30M QA pairs, spanning outdoor/indoor/tabletop scenes and supporting complex reasoning processes (up to 9 steps). We further present TraceSpatial-Bench, a challenging benchmark filling the gap to evaluate spatial tracing. Experimental results show that RoboTracer surpasses baselines in spatial understanding, measuring, and referring, with an average success rate of 79.1%, and also achieves SOTA performance on TraceSpatial-Bench by a large margin, exceeding Gemini-2.5-Pro by 36% accuracy. Notably, RoboTracer can be integrated with various control policies to execute long-horizon, dynamic tasks across diverse robots (UR5, G1 humanoid) in cluttered real-world scenes.</p> + https://arxiv.org/abs/2512.13660 + Mon, 15 Dec 2025 18:52:43 +0000 - VideoVLA: Video Generators Can Be Generalizable Robot Manipulators - https://arxiv.org/abs/2512.06963 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.06963.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yichao Shen, Fangyun Wei, Zhiying Du, Yaobo Liang, Yan Lu, Jiaolong Yang, Nanning Zheng, Baining Guo</p><p><b>Upvotes:</b> 3</p><p><b>Summary:</b> Generalization in robot manipulation is essential for deploying robots in open-world environments and advancing toward artificial general intelligence. While recent Vision-Language-Action (VLA) models leverage large pre-trained understanding models for perception and instruction following, their ability to generalize to novel tasks, objects, and settings remains limited. In this work, we present VideoVLA, a simple approach that explores the potential of transforming large video generation models into robotic VLA manipulators. Given a language instruction and an image, VideoVLA predicts an action sequence as well as the future visual outcomes. Built on a multi-modal Diffusion Transformer, VideoVLA jointly models video, language, and action modalities, using pre-trained video generative models for joint visual and action forecasting. Our experiments show that high-quality imagined futures correlate with reliable action predictions and task success, highlighting the importance of visual imagination in manipulation. VideoVLA demonstrates strong generalization, including imitating other embodiments' skills and handling novel objects. This dual-prediction strategy - forecasting both actions and their visual consequences - explores a paradigm shift in robot learning and unlocks generalization capabilities in manipulation systems.</p> - https://arxiv.org/abs/2512.06963 - Sun, 07 Dec 2025 18:57:15 +0000 + Towards Interactive Intelligence for Digital Humans + https://arxiv.org/abs/2512.13674 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.13674.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yiyi Cai, Xuangeng Chu, Xiwei Gao, Sitong Gong, Yifei Huang, Caixin Kang, Kunhang Li, Haiyang Liu, Ruicong Liu, Yun Liu, Dianwen Ng, Zixiong Su, Erwin Wu, Yuhan Wu, Dingkun Yan, Tianyu Yan, Chang Zeng, Bo Zheng, You Zhou</p><p><b>Upvotes:</b> 9</p><p><b>Summary:</b> We introduce Interactive Intelligence, a novel paradigm of digital human that is capable of personality-aligned expression, adaptive interaction, and self-evolution. To realize this, we present Mio (Multimodal Interactive Omni-Avatar), an end-to-end framework composed of five specialized modules: Thinker, Talker, Face Animator, Body Animator, and Renderer. This unified architecture integrates cognitive reasoning with real-time multimodal embodiment to enable fluid, consistent interaction. Furthermore, we establish a new benchmark to rigorously evaluate the capabilities of interactive intelligence. Extensive experiments demonstrate that our framework achieves superior performance compared to state-of-the-art methods across all evaluated dimensions. Together, these contributions move digital humans beyond superficial imitation toward intelligent interaction.</p> + https://arxiv.org/abs/2512.13674 + Mon, 15 Dec 2025 18:57:35 +0000 - COREA: Coarse-to-Fine 3D Representation Alignment Between Relightable 3D Gaussians and SDF via Bidirectional 3D-to-3D Supervision - https://arxiv.org/abs/2512.07107 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.07107.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jaeyoon Lee, Hojoon Jung, Sungtae Hwang, Jihyong Oh, Jongwon Choi</p><p><b>Upvotes:</b> 8</p><p><b>Summary:</b> We present COREA, the first unified framework that jointly learns relightable 3D Gaussians and a Signed Distance Field (SDF) for accurate geometry reconstruction and faithful relighting. While recent 3D Gaussian Splatting (3DGS) methods have extended toward mesh reconstruction and physically-based rendering (PBR), their geometry is still learned from 2D renderings, leading to coarse surfaces and unreliable BRDF-lighting decomposition. To address these limitations, COREA introduces a coarse-to-fine bidirectional 3D-to-3D alignment strategy that allows geometric signals to be learned directly in 3D space. Within this strategy, depth provides coarse alignment between the two representations, while depth gradients and normals refine fine-scale structure, and the resulting geometry supports stable BRDF-lighting decomposition. A density-control mechanism further stabilizes Gaussian growth, balancing geometric fidelity with memory efficiency. Experiments on standard benchmarks demonstrate that COREA achieves superior performance in novel-view synthesis, mesh reconstruction, and PBR within a unified framework.</p> - https://arxiv.org/abs/2512.07107 - Mon, 08 Dec 2025 02:41:42 +0000 + Feedforward 3D Editing via Text-Steerable Image-to-3D + https://arxiv.org/abs/2512.13678 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.13678.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Ziqi Ma, Hongqiao Chen, Yisong Yue, Georgia Gkioxari</p><p><b>Upvotes:</b> 13</p><p><b>Summary:</b> Recent progress in image-to-3D has opened up immense possibilities for design, AR/VR, and robotics. However, to use AI-generated 3D assets in real applications, a critical requirement is the capability to edit them easily. We present a feedforward method, Steer3D, to add text steerability to image-to-3D models, which enables editing of generated 3D assets with language. Our approach is inspired by ControlNet, which we adapt to image-to-3D generation to enable text steering directly in a forward pass. We build a scalable data engine for automatic data generation, and develop a two-stage training recipe based on flow-matching training and Direct Preference Optimization (DPO). Compared to competing methods, Steer3D more faithfully follows the language instruction and maintains better consistency with the original 3D asset, while being 2.4x to 28.5x faster. Steer3D demonstrates that it is possible to add a new modality (text) to steer the generation of pretrained image-to-3D generative models with 100k data. Project website: https://glab-caltech.github.io/steer3d/</p> + https://arxiv.org/abs/2512.13678 + Mon, 15 Dec 2025 18:58:55 +0000 - SUCCESS-GS: Survey of Compactness and Compression for Efficient Static and Dynamic Gaussian Splatting - https://arxiv.org/abs/2512.07197 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.07197.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Seokhyun Youn, Soohyun Lee, Geonho Kim, Weeyoung Kwon, Sung-Ho Bae, Jihyong Oh</p><p><b>Upvotes:</b> 5</p><p><b>Summary:</b> 3D Gaussian Splatting (3DGS) has emerged as a powerful explicit representation enabling real-time, high-fidelity 3D reconstruction and novel view synthesis. However, its practical use is hindered by the massive memory and computational demands required to store and render millions of Gaussians. These challenges become even more severe in 4D dynamic scenes. To address these issues, the field of Efficient Gaussian Splatting has rapidly evolved, proposing methods that reduce redundancy while preserving reconstruction quality. This survey provides the first unified overview of efficient 3D and 4D Gaussian Splatting techniques. For both 3D and 4D settings, we systematically categorize existing methods into two major directions, Parameter Compression and Restructuring Compression, and comprehensively summarize the core ideas and methodological trends within each category. We further cover widely used datasets, evaluation metrics, and representative benchmark comparisons. Finally, we discuss current limitations and outline promising research directions toward scalable, compact, and real-time Gaussian Splatting for both static and dynamic 3D scene representation.</p> - https://arxiv.org/abs/2512.07197 - Mon, 08 Dec 2025 06:15:59 +0000 + Towards Scalable Pre-training of Visual Tokenizers for Generation + https://arxiv.org/abs/2512.13687 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.13687.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jingfeng Yao, Yuda Song, Yucong Zhou, Xinggang Wang</p><p><b>Upvotes:</b> 89</p><p><b>Summary:</b> The quality of the latent space in visual tokenizers (e.g., VAEs) is crucial for modern generative models. However, the standard reconstruction-based training paradigm produces a latent space that is biased towards low-level information, leading to a foundation flaw: better pixel-level accuracy does not lead to higher-quality generation. This implies that pouring extensive compute into visual tokenizer pre-training translates poorly to improved performance in generation. We identify this as the ``pre-training scaling problem`` and suggest a necessary shift: to be effective for generation, a latent space must concisely represent high-level semantics. We present VTP, a unified visual tokenizer pre-training framework, pioneering the joint optimization of image-text contrastive, self-supervised, and reconstruction losses. Our large-scale study reveals two principal findings: (1) understanding is a key driver of generation, and (2) much better scaling properties, where generative performance scales effectively with compute, parameters, and data allocated to the pretraining of the visual tokenizer. After large-scale pre-training, our tokenizer delivers a competitive profile (78.2 zero-shot accuracy and 0.36 rFID on ImageNet) and 4.1 times faster convergence on generation compared to advanced distillation methods. More importantly, it scales effectively: without modifying standard DiT training specs, solely investing more FLOPS in pretraining VTP achieves 65.8\% FID improvement in downstream generation, while conventional autoencoder stagnates very early at 1/10 FLOPS. Our pre-trained models are available at https://github.com/MiniMax-AI/VTP.</p> + https://arxiv.org/abs/2512.13687 + Mon, 15 Dec 2025 18:59:54 +0000 - Native Parallel Reasoner: Reasoning in Parallelism via Self-Distilled Reinforcement Learning - https://arxiv.org/abs/2512.07461 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.07461.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Tong Wu, Yang Liu, Jun Bai, Zixia Jia, Shuyi Zhang, Ziyong Lin, Yanting Wang, Song-Chun Zhu, Zilong Zheng</p><p><b>Upvotes:</b> 71</p><p><b>Summary:</b> We introduce Native Parallel Reasoner (NPR), a teacher-free framework that enables Large Language Models (LLMs) to self-evolve genuine parallel reasoning capabilities. NPR transforms the model from sequential emulation to native parallel cognition through three key innovations: 1) a self-distilled progressive training paradigm that transitions from ``cold-start'' format discovery to strict topological constraints without external supervision; 2) a novel Parallel-Aware Policy Optimization (PAPO) algorithm that optimizes branching policies directly within the execution graph, allowing the model to learn adaptive decomposition via trial and error; and 3) a robust NPR Engine that refactors memory management and flow control of SGLang to enable stable, large-scale parallel RL training. Across eight reasoning benchmarks, NPR trained on Qwen3-4B achieves performance gains of up to 24.5% and inference speedups up to 4.6x. Unlike prior baselines that often fall back to autoregressive decoding, NPR demonstrates 100% genuine parallel execution, establishing a new standard for self-evolving, efficient, and scalable agentic reasoning.</p> - https://arxiv.org/abs/2512.07461 - Mon, 08 Dec 2025 11:39:43 +0000 + LitePT: Lighter Yet Stronger Point Transformer + https://arxiv.org/abs/2512.13689 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.13689.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yuanwen Yue, Damien Robert, Jianyuan Wang, Sunghwan Hong, Jan Dirk Wegner, Christian Rupprecht, Konrad Schindler</p><p><b>Upvotes:</b> 6</p><p><b>Summary:</b> Modern neural architectures for 3D point cloud processing contain both convolutional layers and attention blocks, but the best way to assemble them remains unclear. We analyse the role of different computational blocks in 3D point cloud networks and find an intuitive behaviour: convolution is adequate to extract low-level geometry at high-resolution in early layers, where attention is expensive without bringing any benefits; attention captures high-level semantics and context in low-resolution, deep layers more efficiently. Guided by this design principle, we propose a new, improved 3D point cloud backbone that employs convolutions in early stages and switches to attention for deeper layers. To avoid the loss of spatial layout information when discarding redundant convolution layers, we introduce a novel, training-free 3D positional encoding, PointROPE. The resulting LitePT model has 3.6times fewer parameters, runs 2times faster, and uses 2times less memory than the state-of-the-art Point Transformer V3, but nonetheless matches or even outperforms it on a range of tasks and datasets. Code and models are available at: https://github.com/prs-eth/LitePT.</p> + https://arxiv.org/abs/2512.13689 + Mon, 15 Dec 2025 18:59:57 +0000 - Unified Video Editing with Temporal Reasoner - https://arxiv.org/abs/2512.07469 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.07469.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Xiangpeng Yang, Ji Xie, Yiyuan Yang, Yan Huang, Min Xu, Qiang Wu</p><p><b>Upvotes:</b> 43</p><p><b>Summary:</b> Existing video editing methods face a critical trade-off: expert models offer precision but rely on task-specific priors like masks, hindering unification; conversely, unified temporal in-context learning models are mask-free but lack explicit spatial cues, leading to weak instruction-to-region mapping and imprecise localization. To resolve this conflict, we propose VideoCoF, a novel Chain-of-Frames approach inspired by Chain-of-Thought reasoning. VideoCoF enforces a ``see, reason, then edit" procedure by compelling the video diffusion model to first predict reasoning tokens (edit-region latents) before generating the target video tokens. This explicit reasoning step removes the need for user-provided masks while achieving precise instruction-to-region alignment and fine-grained video editing. Furthermore, we introduce a RoPE alignment strategy that leverages these reasoning tokens to ensure motion alignment and enable length extrapolation beyond the training duration. We demonstrate that with a minimal data cost of only 50k video pairs, VideoCoF achieves state-of-the-art performance on VideoCoF-Bench, validating the efficiency and effectiveness of our approach. Our code, weight, data are available at https://github.com/knightyxp/VideoCoF.</p> - https://arxiv.org/abs/2512.07469 - Mon, 08 Dec 2025 11:50:18 +0000 + SAGE: Training Smart Any-Horizon Agents for Long Video Reasoning with Reinforcement Learning + https://arxiv.org/abs/2512.13874 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.13874.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jitesh Jain, Jialuo Li, Zixian Ma, Jieyu Zhang, Chris Dongjoo Kim, Sangho Lee, Rohun Tripathi, Tanmay Gupta, Christopher Clark, Humphrey Shi</p><p><b>Upvotes:</b> 15</p><p><b>Summary:</b> As humans, we are natural any-horizon reasoners, i.e., we can decide whether to iteratively skim long videos or watch short ones in full when necessary for a given task. With this in mind, one would expect video reasoning models to reason flexibly across different durations. However, SOTA models are still trained to predict answers in a single turn while processing a large number of frames, akin to watching an entire long video, requiring significant resources. This raises the question: Is it possible to develop performant any-horizon video reasoning systems? Inspired by human behavior, we first propose SAGE, an agent system that performs multi-turn reasoning on long videos while handling simpler problems in a single turn. Secondly, we introduce an easy synthetic data generation pipeline using Gemini-2.5-Flash to train the orchestrator, SAGE-MM, which lies at the core of SAGE. We further propose an effective RL post-training recipe essential for instilling any-horizon reasoning ability in SAGE-MM. Thirdly, we curate SAGE-Bench with an average duration of greater than 700 seconds for evaluating video reasoning ability in real-world entertainment use cases. Lastly, we empirically validate the effectiveness of our system, data, and RL recipe, observing notable improvements of up to 6.1% on open-ended video reasoning tasks, as well as an impressive 8.2% improvement on videos longer than 10 minutes.</p> + https://arxiv.org/abs/2512.13874 + Mon, 15 Dec 2025 20:14:19 +0000 - Beyond Real: Imaginary Extension of Rotary Position Embeddings for Long-Context LLMs - https://arxiv.org/abs/2512.07525 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.07525.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Xiaoran Liu, Yuerong Song, Zhigeng Liu, Zengfeng Huang, Qipeng Guo, Zhaoxiang Liu, Shiguo Lian, Ziwei He, Xipeng Qiu</p><p><b>Upvotes:</b> 55</p><p><b>Summary:</b> Rotary Position Embeddings (RoPE) have become a standard for encoding sequence order in Large Language Models (LLMs) by applying rotations to query and key vectors in the complex plane. Standard implementations, however, utilize only the real component of the complex-valued dot product for attention score calculation. This simplification discards the imaginary component, which contains valuable phase information, leading to a potential loss of relational details crucial for modeling long-context dependencies. In this paper, we propose an extension that re-incorporates this discarded imaginary component. Our method leverages the full complex-valued representation to create a dual-component attention score. We theoretically and empirically demonstrate that this approach enhances the modeling of long-context dependencies by preserving more positional information. Furthermore, evaluations on a suite of long-context language modeling benchmarks show that our method consistently improves performance over the standard RoPE, with the benefits becoming more significant as context length increases. The code is available at https://github.com/OpenMOSS/rope_pp.</p> - https://arxiv.org/abs/2512.07525 - Mon, 08 Dec 2025 12:59:54 +0000 + FiNERweb: Datasets and Artifacts for Scalable Multilingual Named Entity Recognition + https://arxiv.org/abs/2512.13884 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.13884.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jonas Golde, Patrick Haller, Alan Akbik</p><p><b>Upvotes:</b> 14</p><p><b>Summary:</b> Recent multilingual named entity recognition (NER) work has shown that large language models (LLMs) can provide effective synthetic supervision, yet such datasets have mostly appeared as by-products of broader experiments rather than as systematic, reusable resources. We introduce FiNERweb, a dataset-creation pipeline that scales the teacher-student paradigm to 91 languages and 25 scripts. Building on FineWeb-Edu, our approach trains regression models to identify NER-relevant passages and annotates them with multilingual LLMs, resulting in about 225k passages with 235k distinct entity labels. Our experiments show that the regression model achieves more than 84 F1, and that models trained on FiNERweb obtain comparable or improved performance in zero shot transfer settings on English, Thai, and Swahili, despite being trained on 19x less data than strong baselines. In addition, we assess annotation quality using LLM-as-a-judge and observe consistently high scores for both faithfulness (3.99 out of 5) and completeness (4.05 out of 5), indicating reliable and informative annotations. Further, we release the dataset with both English labels and translated label sets in the respective target languages because we observe that the performance of current state-of-the-art models drops by 0.02 to 0.09 F1 when evaluated using target language labels instead of English ones. We release FiNERweb together with all accompanying artifacts to the research community in order to facilitate more effective student-teacher training for multilingual named entity recognition.</p> + https://arxiv.org/abs/2512.13884 + Mon, 15 Dec 2025 20:36:39 +0000 - LongCat-Image Technical Report - https://arxiv.org/abs/2512.07584 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.07584.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Meituan LongCat Team, Hanghang Ma, Haoxian Tan, Jiale Huang, Junqiang Wu, Jun-Yan He, Lishuai Gao, Songlin Xiao, Xiaoming Wei, Xiaoqi Ma, Xunliang Cai, Yayong Guan, Jie Hu</p><p><b>Upvotes:</b> 17</p><p><b>Summary:</b> We introduce LongCat-Image, a pioneering open-source and bilingual (Chinese-English) foundation model for image generation, designed to address core challenges in multilingual text rendering, photorealism, deployment efficiency, and developer accessibility prevalent in current leading models. 1) We achieve this through rigorous data curation strategies across the pre-training, mid-training, and SFT stages, complemented by the coordinated use of curated reward models during the RL phase. This strategy establishes the model as a new state-of-the-art (SOTA), delivering superior text-rendering capabilities and remarkable photorealism, and significantly enhancing aesthetic quality. 2) Notably, it sets a new industry standard for Chinese character rendering. By supporting even complex and rare characters, it outperforms both major open-source and commercial solutions in coverage, while also achieving superior accuracy. 3) The model achieves remarkable efficiency through its compact design. With a core diffusion model of only 6B parameters, it is significantly smaller than the nearly 20B or larger Mixture-of-Experts (MoE) architectures common in the field. This ensures minimal VRAM usage and rapid inference, significantly reducing deployment costs. Beyond generation, LongCat-Image also excels in image editing, achieving SOTA results on standard benchmarks with superior editing consistency compared to other open-source works. 4) To fully empower the community, we have established the most comprehensive open-source ecosystem to date. We are releasing not only multiple model versions for text-to-image and image editing, including checkpoints after mid-training and post-training stages, but also the entire toolchain of training procedure. We believe that the openness of LongCat-Image will provide robust support for developers and researchers, pushing the frontiers of visual content creation.</p> - https://arxiv.org/abs/2512.07584 - Mon, 08 Dec 2025 14:26:40 +0000 + Olmo 3 + https://arxiv.org/abs/2512.13961 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.13961.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Team Olmo, Allyson Ettinger, Amanda Bertsch, Bailey Kuehl, David Graham, David Heineman, Dirk Groeneveld, Faeze Brahman, Finbarr Timbers, Hamish Ivison, Jacob Morrison, Jake Poznanski, Kyle Lo, Luca Soldaini, Matt Jordan, Mayee Chen, Michael Noukhovitch, Nathan Lambert, Pete Walsh, Pradeep Dasigi, Robert Berry, Saumya Malik, Saurabh Shah, Scott Geng, Shane Arora, Shashank Gupta, Taira Anderson, Teng Xiao, Tyler Murray, Tyler Romero, Victoria Graf, Akari Asai, Akshita Bhagia, Alexander Wettig, Alisa Liu, Aman Rangapur, Chloe Anastasiades, Costa Huang, Dustin Schwenk, Harsh Trivedi, Ian Magnusson, Jaron Lochner, Jiacheng Liu, Lester James V. Miranda, Maarten Sap, Malia Morgan, Michael Schmitz, Michal Guerquin, Michael Wilson, Regan Huff, Ronan Le Bras, Rui Xin, Rulin Shao, Sam Skjonsberg, Shannon Zejiang Shen, Shuyue Stella Li, Tucker Wilde, Valentina Pyatkin, Will Merrill, Yapei Chang, Yuling Gu, Zhiyuan Zeng, Ashish Sabharwal, Luke Zettlemoyer, Pang Wei Koh, Ali Farhadi, Noah A. Smith, Hannaneh Hajishirzi</p><p><b>Upvotes:</b> 22</p><p><b>Summary:</b> We introduce Olmo 3, a family of state-of-the-art, fully-open language models at the 7B and 32B parameter scales. Olmo 3 model construction targets long-context reasoning, function calling, coding, instruction following, general chat, and knowledge recall. This release includes the entire model flow, i.e., the full lifecycle of the family of models, including every stage, checkpoint, data point, and dependency used to build it. Our flagship model, Olmo 3 Think 32B, is the strongest fully-open thinking model released to-date.</p> + https://arxiv.org/abs/2512.13961 + Mon, 15 Dec 2025 23:41:48 +0000 - DeepCode: Open Agentic Coding - https://arxiv.org/abs/2512.07921 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.07921.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zongwei Li, Zhonghang Li, Zirui Guo, Xubin Ren, Chao Huang</p><p><b>Upvotes:</b> 19</p><p><b>Summary:</b> Recent advances in large language models (LLMs) have given rise to powerful coding agents, making it possible for code assistants to evolve into code engineers. However, existing methods still face significant challenges in achieving high-fidelity document-to-codebase synthesis--such as scientific papers to code--primarily due to a fundamental conflict between information overload and the context bottlenecks of LLMs. In this work, we introduce DeepCode, a fully autonomous framework that fundamentally addresses this challenge through principled information-flow management. By treating repository synthesis as a channel optimization problem, DeepCode seamlessly orchestrates four information operations to maximize task-relevant signals under finite context budgets: source compression via blueprint distillation, structured indexing using stateful code memory, conditional knowledge injection via retrieval-augmented generation, and closed-loop error correction. Extensive evaluations on the PaperBench benchmark demonstrate that DeepCode achieves state-of-the-art performance, decisively outperforming leading commercial agents such as Cursor and Claude Code, and crucially, surpassing PhD-level human experts from top institutes on key reproduction metrics. By systematically transforming paper specifications into production-grade implementations comparable to human expert quality, this work establishes new foundations for autonomous scientific reproduction that can accelerate research evaluation and discovery.</p> - https://arxiv.org/abs/2512.07921 - Mon, 08 Dec 2025 16:07:13 +0000 + Sparse-LaViDa: Sparse Multimodal Discrete Diffusion Language Models + https://arxiv.org/abs/2512.14008 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.14008.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Shufan Li, Jiuxiang Gu, Kangning Liu, Zhe Lin, Zijun Wei, Aditya Grover, Jason Kuen</p><p><b>Upvotes:</b> 8</p><p><b>Summary:</b> Masked Discrete Diffusion Models (MDMs) have achieved strong performance across a wide range of multimodal tasks, including image understanding, generation, and editing. However, their inference speed remains suboptimal due to the need to repeatedly process redundant masked tokens at every sampling step. In this work, we propose Sparse-LaViDa, a novel modeling framework that dynamically truncates unnecessary masked tokens at each inference step to accelerate MDM sampling. To preserve generation quality, we introduce specialized register tokens that serve as compact representations for the truncated tokens. Furthermore, to ensure consistency between training and inference, we design a specialized attention mask that faithfully matches the truncated sampling procedure during training. Built upon the state-of-the-art unified MDM LaViDa-O, Sparse-LaViDa achieves up to a 2x speedup across diverse tasks including text-to-image generation, image editing, and mathematical reasoning, while maintaining generation quality.</p> + https://arxiv.org/abs/2512.14008 + Tue, 16 Dec 2025 02:06:06 +0000 - Distribution Matching Variational AutoEncoder - https://arxiv.org/abs/2512.07778 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.07778.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Sen Ye, Jianning Pei, Mengde Xu, Shuyang Gu, Chunyu Wang, Liwei Wang, Han Hu</p><p><b>Upvotes:</b> 26</p><p><b>Summary:</b> Most visual generative models compress images into a latent space before applying diffusion or autoregressive modelling. Yet, existing approaches such as VAEs and foundation model aligned encoders implicitly constrain the latent space without explicitly shaping its distribution, making it unclear which types of distributions are optimal for modeling. We introduce Distribution-Matching VAE (DMVAE), which explicitly aligns the encoder's latent distribution with an arbitrary reference distribution via a distribution matching constraint. This generalizes beyond the Gaussian prior of conventional VAEs, enabling alignment with distributions derived from self-supervised features, diffusion noise, or other prior distributions. With DMVAE, we can systematically investigate which latent distributions are more conducive to modeling, and we find that SSL-derived distributions provide an excellent balance between reconstruction fidelity and modeling efficiency, reaching gFID equals 3.2 on ImageNet with only 64 training epochs. Our results suggest that choosing a suitable latent distribution structure (achieved via distribution-level alignment), rather than relying on fixed priors, is key to bridging the gap between easy-to-model latents and high-fidelity image synthesis. Code is avaliable at https://github.com/sen-ye/dmvae.</p> - https://arxiv.org/abs/2512.07778 - Mon, 08 Dec 2025 17:59:47 +0000 + OpenDataArena: A Fair and Open Arena for Benchmarking Post-Training Dataset Value + https://arxiv.org/abs/2512.14051 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.14051.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Mengzhang Cai, Xin Gao, Yu Li, Honglin Lin, Zheng Liu, Zhuoshi Pan, Qizhi Pei, Xiaoran Shang, Mengyuan Sun, Zinan Tang, Xiaoyang Wang, Zhanping Zhong, Yun Zhu, Dahua Lin, Conghui He, Lijun Wu</p><p><b>Upvotes:</b> 37</p><p><b>Summary:</b> The rapid evolution of Large Language Models (LLMs) is predicated on the quality and diversity of post-training datasets. However, a critical dichotomy persists: while models are rigorously benchmarked, the data fueling them remains a black box--characterized by opaque composition, uncertain provenance, and a lack of systematic evaluation. This opacity hinders reproducibility and obscures the causal link between data characteristics and model behaviors. To bridge this gap, we introduce OpenDataArena (ODA), a holistic and open platform designed to benchmark the intrinsic value of post-training data. ODA establishes a comprehensive ecosystem comprising four key pillars: (i) a unified training-evaluation pipeline that ensures fair, open comparisons across diverse models (e.g., Llama, Qwen) and domains; (ii) a multi-dimensional scoring framework that profiles data quality along tens of distinct axes; (iii) an interactive data lineage explorer to visualize dataset genealogy and dissect component sources; and (iv) a fully open-source toolkit for training, evaluation, and scoring to foster data research. Extensive experiments on ODA--covering over 120 training datasets across multiple domains on 22 benchmarks, validated by more than 600 training runs and 40 million processed data points--reveal non-trivial insights. Our analysis uncovers the inherent trade-offs between data complexity and task performance, identifies redundancy in popular benchmarks through lineage tracing, and maps the genealogical relationships across datasets. We release all results, tools, and configurations to democratize access to high-quality data evaluation. Rather than merely expanding a leaderboard, ODA envisions a shift from trial-and-error data curation to a principled science of Data-Centric AI, paving the way for rigorous studies on data mixing laws and the strategic composition of foundation models.</p> + https://arxiv.org/abs/2512.14051 + Tue, 16 Dec 2025 03:33:24 +0000 - On the Interplay of Pre-Training, Mid-Training, and RL on Reasoning Language Models - https://arxiv.org/abs/2512.07783 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.07783.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Charlie Zhang, Graham Neubig, Xiang Yue</p><p><b>Upvotes:</b> 31</p><p><b>Summary:</b> Recent reinforcement learning (RL) techniques have yielded impressive reasoning improvements in language models, yet it remains unclear whether post-training truly extends a model's reasoning ability beyond what it acquires during pre-training. A central challenge is the lack of control in modern training pipelines: large-scale pre-training corpora are opaque, mid-training is often underexamined, and RL objectives interact with unknown prior knowledge in complex ways. To resolve this ambiguity, we develop a fully controlled experimental framework that isolates the causal contributions of pre-training, mid-training, and RL-based post-training. Our approach employs synthetic reasoning tasks with explicit atomic operations, parseable step-by-step reasoning traces, and systematic manipulation of training distributions. We evaluate models along two axes: extrapolative generalization to more complex compositions and contextual generalization across surface contexts. Using this framework, we reconcile competing views on RL's effectiveness. We show that: 1) RL produces true capability gains (pass@128) only when pre-training leaves sufficient headroom and when RL data target the model's edge of competence, tasks at the boundary that are difficult but not yet out of reach. 2) Contextual generalization requires minimal yet sufficient pre-training exposure, after which RL can reliably transfer. 3) Mid-training significantly enhances performance under fixed compute compared with RL only, demonstrating its central but underexplored role in training pipelines. 4) Process-level rewards reduce reward hacking and improve reasoning fidelity. Together, these results clarify the interplay between pre-training, mid-training, and RL, offering a foundation for understanding and improving reasoning LM training strategies.</p> - https://arxiv.org/abs/2512.07783 - Mon, 08 Dec 2025 18:12:10 +0000 + HyperVL: An Efficient and Dynamic Multimodal Large Language Model for Edge Devices + https://arxiv.org/abs/2512.14052 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.14052.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> HyperAI Team, Yuchen Liu, Kaiyang Han, Zhiqiang Xia, Yuhang Dong, Chen Song, Kangyu Tang, Jiaming Xu, Xiushi Feng, WenXuan Yu, Li Peng, Mingyang Wang, Kai Wang, Changpeng Yang, Yang Li, Haoyu Lu, Hao Wang, Bingna Xu, Guangyao Liu, Long Huang, Kaibin Guo, Jinyang Wu, Dan Wu, Hongzhen Wang, Peng Zhou, Shuai Nie, Shande Wang, Runyu Shi, Ying Huang</p><p><b>Upvotes:</b> 39</p><p><b>Summary:</b> Current multimodal large lanauge models possess strong perceptual and reasoning capabilities, however high computational and memory requirements make them difficult to deploy directly on on-device environments. While small-parameter models are progressively endowed with strong general capabilities, standard Vision Transformer (ViT) encoders remain a critical bottleneck, suffering from excessive latency and memory consumption when processing high-resolution inputs.To address these challenges, we introduce HyperVL, an efficient multimodal large language model tailored for on-device inference. HyperVL adopts an image-tiling strategy to cap peak memory usage and incorporates two novel techniques: (1) a Visual Resolution Compressor (VRC) that adaptively predicts optimal encoding resolutions to eliminate redundant computation, and (2) Dual Consistency Learning (DCL), which aligns multi-scale ViT encoders within a unified framework, enabling dynamic switching between visual branches under a shared LLM. Extensive experiments demonstrate that HyperVL achieves state-of-the-art performance among models of comparable size across multiple benchmarks. Furthermore, it significantly significantly reduces latency and power consumption on real mobile devices, demonstrating its practicality for on-device multimodal inference.</p> + https://arxiv.org/abs/2512.14052 + Tue, 16 Dec 2025 03:36:41 +0000 - OneStory: Coherent Multi-Shot Video Generation with Adaptive Memory - https://arxiv.org/abs/2512.07802 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.07802.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zhaochong An, Menglin Jia, Haonan Qiu, Zijian Zhou, Xiaoke Huang, Zhiheng Liu, Weiming Ren, Kumara Kahatapitiya, Ding Liu, Sen He, Chenyang Zhang, Tao Xiang, Fanny Yang, Serge Belongie, Tian Xie</p><p><b>Upvotes:</b> 43</p><p><b>Summary:</b> Storytelling in real-world videos often unfolds through multiple shots -- discontinuous yet semantically connected clips that together convey a coherent narrative. However, existing multi-shot video generation (MSV) methods struggle to effectively model long-range cross-shot context, as they rely on limited temporal windows or single keyframe conditioning, leading to degraded performance under complex narratives. In this work, we propose OneStory, enabling global yet compact cross-shot context modeling for consistent and scalable narrative generation. OneStory reformulates MSV as a next-shot generation task, enabling autoregressive shot synthesis while leveraging pretrained image-to-video (I2V) models for strong visual conditioning. We introduce two key modules: a Frame Selection module that constructs a semantically-relevant global memory based on informative frames from prior shots, and an Adaptive Conditioner that performs importance-guided patchification to generate compact context for direct conditioning. We further curate a high-quality multi-shot dataset with referential captions to mirror real-world storytelling patterns, and design effective training strategies under the next-shot paradigm. Finetuned from a pretrained I2V model on our curated 60K dataset, OneStory achieves state-of-the-art narrative coherence across diverse and complex scenes in both text- and image-conditioned settings, enabling controllable and immersive long-form video storytelling.</p> - https://arxiv.org/abs/2512.07802 - Mon, 08 Dec 2025 18:32:24 +0000 + Efficient-DLM: From Autoregressive to Diffusion Language Models, and Beyond in Speed + https://arxiv.org/abs/2512.14067 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.14067.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yonggan Fu, Lexington Whalen, Zhifan Ye, Xin Dong, Shizhe Diao, Jingyu Liu, Chengyue Wu, Hao Zhang, Enze Xie, Song Han, Maksim Khadkevich, Jan Kautz, Yingyan Celine Lin, Pavlo Molchanov</p><p><b>Upvotes:</b> 11</p><p><b>Summary:</b> Diffusion language models (dLMs) have emerged as a promising paradigm that enables parallel, non-autoregressive generation, but their learning efficiency lags behind that of autoregressive (AR) language models when trained from scratch. To this end, we study AR-to-dLM conversion to transform pretrained AR models into efficient dLMs that excel in speed while preserving AR models' task accuracy. We achieve this by identifying limitations in the attention patterns and objectives of existing AR-to-dLM methods and then proposing principles and methodologies for more effective AR-to-dLM conversion. Specifically, we first systematically compare different attention patterns and find that maintaining pretrained AR weight distributions is critical for effective AR-to-dLM conversion. As such, we introduce a continuous pretraining scheme with a block-wise attention pattern, which remains causal across blocks while enabling bidirectional modeling within each block. We find that this approach can better preserve pretrained AR models' weight distributions than fully bidirectional modeling, in addition to its known benefit of enabling KV caching, and leads to a win-win in accuracy and efficiency. Second, to mitigate the training-test gap in mask token distributions (uniform vs. highly left-to-right), we propose a position-dependent token masking strategy that assigns higher masking probabilities to later tokens during training to better mimic test-time behavior. Leveraging this framework, we conduct extensive studies of dLMs' attention patterns, training dynamics, and other design choices, providing actionable insights into scalable AR-to-dLM conversion. These studies lead to the Efficient-DLM family, which outperforms state-of-the-art AR models and dLMs, e.g., our Efficient-DLM 8B achieves +5.4%/+2.7% higher accuracy with 4.5x/2.7x higher throughput compared to Dream 7B and Qwen3 4B, respectively.</p> + https://arxiv.org/abs/2512.14067 + Tue, 16 Dec 2025 04:12:17 +0000 - Group Representational Position Encoding - https://arxiv.org/abs/2512.07805 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.07805.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yifan Zhang, Zixiang Chen, Yifeng Liu, Zhen Qin, Huizhuo Yuan, Kangping Xu, Yang Yuan, Quanquan Gu, Andrew Chi-Chih Yao</p><p><b>Upvotes:</b> 3</p><p><b>Summary:</b> We present GRAPE (Group RepresentAtional Position Encoding), a unified framework for positional encoding based on group actions. GRAPE brings together two families of mechanisms: (i) multiplicative rotations (Multiplicative GRAPE) in SO(d) and (ii) additive logit biases (Additive GRAPE) arising from unipotent actions in the general linear group GL. In Multiplicative GRAPE, a position n in Z (or t in R) acts as G(n)=exp(n,ω,L) with a rank-2 skew generator L in R^{d times d}, yielding a relative, compositional, norm-preserving map with a closed-form matrix exponential. RoPE is recovered exactly when the d/2 planes are the canonical coordinate pairs with log-uniform spectrum. Learned commuting subspaces and compact non-commuting mixtures strictly extend this geometry to capture cross-subspace feature coupling at O(d) and O(r d) cost per head, respectively. In Additive GRAPE, additive logits arise as rank-1 (or low-rank) unipotent actions, recovering ALiBi and the Forgetting Transformer (FoX) as exact special cases while preserving an exact relative law and streaming cacheability. Altogether, GRAPE supplies a principled design space for positional geometry in long-context models, subsuming RoPE and ALiBi as special cases. Project Page: https://github.com/model-architectures/GRAPE.</p> - https://arxiv.org/abs/2512.07805 - Mon, 08 Dec 2025 18:39:13 +0000 + Zoom-Zero: Reinforced Coarse-to-Fine Video Understanding via Temporal Zoom-in + https://arxiv.org/abs/2512.14273 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.14273.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Xiaoqian Shen, Min-Hung Chen, Yu-Chiang Frank Wang, Mohamed Elhoseiny, Ryo Hachiuma</p><p><b>Upvotes:</b> 7</p><p><b>Summary:</b> Grounded video question answering (GVQA) aims to localize relevant temporal segments in videos and generate accurate answers to a given question; however, large video-language models (LVLMs) exhibit limited temporal awareness. Although existing approaches based on Group Relative Policy Optimization (GRPO) attempt to improve temporal grounding, they still struggle to faithfully ground their answers in the relevant video evidence, leading to temporal mislocalization and hallucinations. In this work, we present Zoom-Zero, a coarse-to-fine framework that first localizes query-relevant segments and then temporally zooms into the most salient frames for finer-grained visual verification. Our method addresses the limits of GRPO for the GVQA task with two key innovations: (i) a zoom-in accuracy reward that validates the fidelity of temporal grounding prediction and facilitates fine-grained visual verification on grounded frames; (ii) token-selective credit assignment, which attributes rewards to the tokens responsible for temporal localization or answer generation, mitigating GRPO's issue in handling multi-faceted reward signals. Our proposed method advances grounded video question answering, improving temporal grounding by 5.2\% on NExT-GQA and 4.6\% on ReXTime, while also enhancing average answer accuracy by 2.4\%. Additionally, the coarse-to-fine zoom-in during inference further benefits long-form video understanding by preserving critical visual details without compromising global context, yielding an average improvement of 6.4\% on long-video benchmarks.</p> + https://arxiv.org/abs/2512.14273 + Tue, 16 Dec 2025 10:34:39 +0000 - Multi-view Pyramid Transformer: Look Coarser to See Broader - https://arxiv.org/abs/2512.07806 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.07806.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Gyeongjin Kang, Seungkwon Yang, Seungtae Nam, Younggeun Lee, Jungwoo Kim, Eunbyung Park</p><p><b>Upvotes:</b> 20</p><p><b>Summary:</b> We propose Multi-view Pyramid Transformer (MVP), a scalable multi-view transformer architecture that directly reconstructs large 3D scenes from tens to hundreds of images in a single forward pass. Drawing on the idea of ``looking broader to see the whole, looking finer to see the details," MVP is built on two core design principles: 1) a local-to-global inter-view hierarchy that gradually broadens the model's perspective from local views to groups and ultimately the full scene, and 2) a fine-to-coarse intra-view hierarchy that starts from detailed spatial representations and progressively aggregates them into compact, information-dense tokens. This dual hierarchy achieves both computational efficiency and representational richness, enabling fast reconstruction of large and complex scenes. We validate MVP on diverse datasets and show that, when coupled with 3D Gaussian Splatting as the underlying 3D representation, it achieves state-of-the-art generalizable reconstruction quality while maintaining high efficiency and scalability across a wide range of view configurations.</p> - https://arxiv.org/abs/2512.07806 - Mon, 08 Dec 2025 18:39:27 +0000 + SS4D: Native 4D Generative Model via Structured Spacetime Latents + https://arxiv.org/abs/2512.14284 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.14284.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zhibing Li, Mengchen Zhang, Tong Wu, Jing Tan, Jiaqi Wang, Dahua Lin</p><p><b>Upvotes:</b> 11</p><p><b>Summary:</b> We present SS4D, a native 4D generative model that synthesizes dynamic 3D objects directly from monocular video. Unlike prior approaches that construct 4D representations by optimizing over 3D or video generative models, we train a generator directly on 4D data, achieving high fidelity, temporal coherence, and structural consistency. At the core of our method is a compressed set of structured spacetime latents. Specifically, (1) To address the scarcity of 4D training data, we build on a pre-trained single-image-to-3D model, preserving strong spatial consistency. (2) Temporal consistency is enforced by introducing dedicated temporal layers that reason across frames. (3) To support efficient training and inference over long video sequences, we compress the latent sequence along the temporal axis using factorized 4D convolutions and temporal downsampling blocks. In addition, we employ a carefully designed training strategy to enhance robustness against occlusion</p> + https://arxiv.org/abs/2512.14284 + Tue, 16 Dec 2025 10:45:06 +0000 - One Layer Is Enough: Adapting Pretrained Visual Encoders for Image Generation - https://arxiv.org/abs/2512.07829 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.07829.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yuan Gao, Chen Chen, Tianrong Chen, Jiatao Gu</p><p><b>Upvotes:</b> 17</p><p><b>Summary:</b> Visual generative models (e.g., diffusion models) typically operate in compressed latent spaces to balance training efficiency and sample quality. In parallel, there has been growing interest in leveraging high-quality pre-trained visual representations, either by aligning them inside VAEs or directly within the generative model. However, adapting such representations remains challenging due to fundamental mismatches between understanding-oriented features and generation-friendly latent spaces. Representation encoders benefit from high-dimensional latents that capture diverse hypotheses for masked regions, whereas generative models favor low-dimensional latents that must faithfully preserve injected noise. This discrepancy has led prior work to rely on complex objectives and architectures. In this work, we propose FAE (Feature Auto-Encoder), a simple yet effective framework that adapts pre-trained visual representations into low-dimensional latents suitable for generation using as little as a single attention layer, while retaining sufficient information for both reconstruction and understanding. The key is to couple two separate deep decoders: one trained to reconstruct the original feature space, and a second that takes the reconstructed features as input for image generation. FAE is generic; it can be instantiated with a variety of self-supervised encoders (e.g., DINO, SigLIP) and plugged into two distinct generative families: diffusion models and normalizing flows. Across class-conditional and text-to-image benchmarks, FAE achieves strong performance. For example, on ImageNet 256x256, our diffusion model with CFG attains a near state-of-the-art FID of 1.29 (800 epochs) and 1.70 (80 epochs). Without CFG, FAE reaches the state-of-the-art FID of 1.48 (800 epochs) and 2.08 (80 epochs), demonstrating both high quality and fast learning.</p> - https://arxiv.org/abs/2512.07829 - Mon, 08 Dec 2025 18:57:26 +0000 + Vector Prism: Animating Vector Graphics by Stratifying Semantic Structure + https://arxiv.org/abs/2512.14336 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.14336.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jooyeol Yun, Jaegul Choo</p><p><b>Upvotes:</b> 27</p><p><b>Summary:</b> Scalable Vector Graphics (SVG) are central to modern web design, and the demand to animate them continues to grow as web environments become increasingly dynamic. Yet automating the animation of vector graphics remains challenging for vision-language models (VLMs) despite recent progress in code generation and motion planning. VLMs routinely mis-handle SVGs, since visually coherent parts are often fragmented into low-level shapes that offer little guidance of which elements should move together. In this paper, we introduce a framework that recovers the semantic structure required for reliable SVG animation and reveals the missing layer that current VLM systems overlook. This is achieved through a statistical aggregation of multiple weak part predictions, allowing the system to stably infer semantics from noisy predictions. By reorganizing SVGs into semantic groups, our approach enables VLMs to produce animations with far greater coherence. Our experiments demonstrate substantial gains over existing approaches, suggesting that semantic recovery is the key step that unlocks robust SVG animation and supports more interpretable interactions between VLMs and vector graphics.</p> + https://arxiv.org/abs/2512.14336 + Tue, 16 Dec 2025 12:03:46 +0000 - UnityVideo: Unified Multi-Modal Multi-Task Learning for Enhancing World-Aware Video Generation - https://arxiv.org/abs/2512.07831 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.07831.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jiehui Huang, Yuechen Zhang, Xu He, Yuan Gao, Zhi Cen, Bin Xia, Yan Zhou, Xin Tao, Pengfei Wan, Jiaya Jia</p><p><b>Upvotes:</b> 16</p><p><b>Summary:</b> Recent video generation models demonstrate impressive synthesis capabilities but remain limited by single-modality conditioning, constraining their holistic world understanding. This stems from insufficient cross-modal interaction and limited modal diversity for comprehensive world knowledge representation. To address these limitations, we introduce UnityVideo, a unified framework for world-aware video generation that jointly learns across multiple modalities (segmentation masks, human skeletons, DensePose, optical flow, and depth maps) and training paradigms. Our approach features two core components: (1) dynamic noising to unify heterogeneous training paradigms, and (2) a modality switcher with an in-context learner that enables unified processing via modular parameters and contextual learning. We contribute a large-scale unified dataset with 1.3M samples. Through joint optimization, UnityVideo accelerates convergence and significantly enhances zero-shot generalization to unseen data. We demonstrate that UnityVideo achieves superior video quality, consistency, and improved alignment with physical world constraints. Code and data can be found at: https://github.com/dvlab-research/UnityVideo</p> - https://arxiv.org/abs/2512.07831 - Mon, 08 Dec 2025 18:59:01 +0000 + RePo: Language Models with Context Re-Positioning + https://arxiv.org/abs/2512.14391 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.14391.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Huayang Li, Tianyu Zhao, Richard Sproat</p><p><b>Upvotes:</b> 7</p><p><b>Summary:</b> In-context learning is fundamental to modern Large Language Models (LLMs); however, prevailing architectures impose a rigid and fixed contextual structure by assigning linear or constant positional indices. Drawing on Cognitive Load Theory (CLT), we argue that this uninformative structure increases extraneous cognitive load, consuming finite working memory capacity that should be allocated to deep reasoning and attention allocation. To address this, we propose RePo, a novel mechanism that reduces extraneous load via context re-positioning. Unlike standard approaches, RePo utilizes a differentiable module, f_φ, to assign token positions that capture contextual dependencies, rather than replying on pre-defined integer range. By continually pre-training on the OLMo-2 1B backbone, we demonstrate that RePo significantly enhances performance on tasks involving noisy contexts, structured data, and longer context length, while maintaining competitive performance on general short-context tasks. Detailed analysis reveals that RePo successfully allocate higher attention to distant but relevant information, assign positions in dense and non-linear space, and capture the intrinsic structure of the input context. Our code is available at https://github.com/SakanaAI/repo.</p> + https://arxiv.org/abs/2512.14391 + Tue, 16 Dec 2025 13:30:30 +0000 - Relational Visual Similarity - https://arxiv.org/abs/2512.07833 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.07833.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Thao Nguyen, Sicheng Mo, Krishna Kumar Singh, Yilin Wang, Jing Shi, Nicholas Kolkin, Eli Shechtman, Yong Jae Lee, Yuheng Li</p><p><b>Upvotes:</b> 23</p><p><b>Summary:</b> Humans do not just see attribute similarity -- we also see relational similarity. An apple is like a peach because both are reddish fruit, but the Earth is also like a peach: its crust, mantle, and core correspond to the peach's skin, flesh, and pit. This ability to perceive and recognize relational similarity, is arguable by cognitive scientist to be what distinguishes humans from other species. Yet, all widely used visual similarity metrics today (e.g., LPIPS, CLIP, DINO) focus solely on perceptual attribute similarity and fail to capture the rich, often surprising relational similarities that humans perceive. How can we go beyond the visible content of an image to capture its relational properties? How can we bring images with the same relational logic closer together in representation space? To answer these questions, we first formulate relational image similarity as a measurable problem: two images are relationally similar when their internal relations or functions among visual elements correspond, even if their visual attributes differ. We then curate 114k image-caption dataset in which the captions are anonymized -- describing the underlying relational logic of the scene rather than its surface content. Using this dataset, we finetune a Vision-Language model to measure the relational similarity between images. This model serves as the first step toward connecting images by their underlying relational structure rather than their visible appearance. Our study shows that while relational similarity has a lot of real-world applications, existing image similarity models fail to capture it -- revealing a critical gap in visual computing.</p> - https://arxiv.org/abs/2512.07833 - Mon, 08 Dec 2025 18:59:56 +0000 + A4-Agent: An Agentic Framework for Zero-Shot Affordance Reasoning + https://arxiv.org/abs/2512.14442 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.14442.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zixin Zhang, Kanghao Chen, Hanqing Wang, Hongfei Zhang, Harold Haodong Chen, Chenfei Liao, Litao Guo, Ying-Cong Chen</p><p><b>Upvotes:</b> 10</p><p><b>Summary:</b> Affordance prediction, which identifies interaction regions on objects based on language instructions, is critical for embodied AI. Prevailing end-to-end models couple high-level reasoning and low-level grounding into a single monolithic pipeline and rely on training over annotated datasets, which leads to poor generalization on novel objects and unseen environments. In this paper, we move beyond this paradigm by proposing A4-Agent, a training-free agentic framework that decouples affordance prediction into a three-stage pipeline. Our framework coordinates specialized foundation models at test time: (1) a Dreamer that employs generative models to visualize how an interaction would look; (2) a Thinker that utilizes large vision-language models to decide what object part to interact with; and (3) a Spotter that orchestrates vision foundation models to precisely locate where the interaction area is. By leveraging the complementary strengths of pre-trained models without any task-specific fine-tuning, our zero-shot framework significantly outperforms state-of-the-art supervised methods across multiple benchmarks and demonstrates robust generalization to real-world settings.</p> + https://arxiv.org/abs/2512.14442 + Tue, 16 Dec 2025 14:27:47 +0000 - Voxify3D: Pixel Art Meets Volumetric Rendering - https://arxiv.org/abs/2512.07834 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.07834.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yi-Chuan Huang, Jiewen Chan, Hao-Jen Chien, Yu-Lun Liu</p><p><b>Upvotes:</b> 39</p><p><b>Summary:</b> Voxel art is a distinctive stylization widely used in games and digital media, yet automated generation from 3D meshes remains challenging due to conflicting requirements of geometric abstraction, semantic preservation, and discrete color coherence. Existing methods either over-simplify geometry or fail to achieve the pixel-precise, palette-constrained aesthetics of voxel art. We introduce Voxify3D, a differentiable two-stage framework bridging 3D mesh optimization with 2D pixel art supervision. Our core innovation lies in the synergistic integration of three components: (1) orthographic pixel art supervision that eliminates perspective distortion for precise voxel-pixel alignment; (2) patch-based CLIP alignment that preserves semantics across discretization levels; (3) palette-constrained Gumbel-Softmax quantization enabling differentiable optimization over discrete color spaces with controllable palette strategies. This integration addresses fundamental challenges: semantic preservation under extreme discretization, pixel-art aesthetics through volumetric rendering, and end-to-end discrete optimization. Experiments show superior performance (37.12 CLIP-IQA, 77.90\% user preference) across diverse characters and controllable abstraction (2-8 colors, 20x-50x resolutions). Project page: https://yichuanh.github.io/Voxify-3D/</p> - https://arxiv.org/abs/2512.07834 - Mon, 08 Dec 2025 18:59:58 +0000 + RecGPT-V2 Technical Report + https://arxiv.org/abs/2512.14503 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.14503.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Chao Yi, Dian Chen, Gaoyang Guo, Jiakai Tang, Jian Wu, Jing Yu, Mao Zhang, Wen Chen, Wenjun Yang, Yujie Luo, Yuning Jiang, Zhujin Gao, Bo Zheng, Binbin Cao, Changfa Wu, Dixuan Wang, Han Wu, Haoyi Hu, Kewei Zhu, Lang Tian, Lin Yang, Qiqi Huang, Siqi Yang, Wenbo Su, Xiaoxiao He, Xin Tong, Xu Chen, Xunke Xi, Xiaowei Huang, Yaxuan Wu, Yeqiu Yang, Yi Hu, Yujin Yuan, Yuliang Yan, Zile Zhou</p><p><b>Upvotes:</b> 16</p><p><b>Summary:</b> Large language models (LLMs) have demonstrated remarkable potential in transforming recommender systems from implicit behavioral pattern matching to explicit intent reasoning. While RecGPT-V1 successfully pioneered this paradigm by integrating LLM-based reasoning into user interest mining and item tag prediction, it suffers from four fundamental limitations: (1) computational inefficiency and cognitive redundancy across multiple reasoning routes; (2) insufficient explanation diversity in fixed-template generation; (3) limited generalization under supervised learning paradigms; and (4) simplistic outcome-focused evaluation that fails to match human standards. To address these challenges, we present RecGPT-V2 with four key innovations. First, a Hierarchical Multi-Agent System restructures intent reasoning through coordinated collaboration, eliminating cognitive duplication while enabling diverse intent coverage. Combined with Hybrid Representation Inference that compresses user-behavior contexts, our framework reduces GPU consumption by 60% and improves exclusive recall from 9.39% to 10.99%. Second, a Meta-Prompting framework dynamically generates contextually adaptive prompts, improving explanation diversity by +7.3%. Third, constrained reinforcement learning mitigates multi-reward conflicts, achieving +24.1% improvement in tag prediction and +13.0% in explanation acceptance. Fourth, an Agent-as-a-Judge framework decomposes assessment into multi-step reasoning, improving human preference alignment. Online A/B tests on Taobao demonstrate significant improvements: +2.98% CTR, +3.71% IPV, +2.19% TV, and +11.46% NER. RecGPT-V2 establishes both the technical feasibility and commercial viability of deploying LLM-powered intent reasoning at scale, bridging the gap between cognitive exploration and industrial utility.</p> + https://arxiv.org/abs/2512.14503 + Tue, 16 Dec 2025 15:40:44 +0000 - Preserving Source Video Realism: High-Fidelity Face Swapping for Cinematic Quality - https://arxiv.org/abs/2512.07951 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.07951.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zekai Luo, Zongze Du, Zhouhang Zhu, Hao Zhong, Muzhi Zhu, Wen Wang, Yuling Xi, Chenchen Jing, Hao Chen, Chunhua Shen</p><p><b>Upvotes:</b> 46</p><p><b>Summary:</b> Video face swapping is crucial in film and entertainment production, where achieving high fidelity and temporal consistency over long and complex video sequences remains a significant challenge. Inspired by recent advances in reference-guided image editing, we explore whether rich visual attributes from source videos can be similarly leveraged to enhance both fidelity and temporal coherence in video face swapping. Building on this insight, this work presents LivingSwap, the first video reference guided face swapping model. Our approach employs keyframes as conditioning signals to inject the target identity, enabling flexible and controllable editing. By combining keyframe conditioning with video reference guidance, the model performs temporal stitching to ensure stable identity preservation and high-fidelity reconstruction across long video sequences. To address the scarcity of data for reference-guided training, we construct a paired face-swapping dataset, Face2Face, and further reverse the data pairs to ensure reliable ground-truth supervision. Extensive experiments demonstrate that our method achieves state-of-the-art results, seamlessly integrating the target identity with the source video's expressions, lighting, and motion, while significantly reducing manual effort in production workflows. Project webpage: https://aim-uofa.github.io/LivingSwap</p> - https://arxiv.org/abs/2512.07951 - Mon, 08 Dec 2025 19:00:04 +0000 + VersatileFFN: Achieving Parameter Efficiency in LLMs via Adaptive Wide-and-Deep Reuse + https://arxiv.org/abs/2512.14531 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.14531.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Ying Nie, Kai Han, Hongguang Li, Hang Zhou, Tianyu Guo, Enhua Wu, Xinghao Chen, Yunhe Wang</p><p><b>Upvotes:</b> 11</p><p><b>Summary:</b> The rapid scaling of Large Language Models (LLMs) has achieved remarkable performance, but it also leads to prohibitive memory costs. Existing parameter-efficient approaches such as pruning and quantization mainly compress pretrained models without enhancing architectural capacity, thereby hitting the representational ceiling of the base model. In this work, we propose VersatileFFN, a novel feed-forward network (FFN) that enables flexible reuse of parameters in both width and depth dimensions within a fixed parameter budget. Inspired by the dual-process theory of cognition, VersatileFFN comprises two adaptive pathways: a width-versatile path that generates a mixture of sub-experts from a single shared FFN, mimicking sparse expert routing without increasing parameters, and a depth-versatile path that recursively applies the same FFN to emulate deeper processing for complex tokens. A difficulty-aware gating dynamically balances the two pathways, steering "easy" tokens through the efficient width-wise route and allocating deeper iterative refinement to "hard" tokens. Crucially, both pathways reuse the same parameters, so all additional capacity comes from computation rather than memory. Experiments across diverse benchmarks and model scales demonstrate the effectiveness of the method. The code will be available at https://github.com/huawei-noah/noah-research/tree/master/VersatileFFN.</p> + https://arxiv.org/abs/2512.14531 + Tue, 16 Dec 2025 16:08:23 +0000 - Beyond Unified Models: A Service-Oriented Approach to Low Latency, Context Aware Phonemization for Real Time TTS - https://arxiv.org/abs/2512.08006 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.08006.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Mahta Fetrat, Donya Navabi, Zahra Dehghanian, Morteza Abolghasemi, Hamid R. Rabiee</p><p><b>Upvotes:</b> 2</p><p><b>Summary:</b> Lightweight, real-time text-to-speech systems are crucial for accessibility. However, the most efficient TTS models often rely on lightweight phonemizers that struggle with context-dependent challenges. In contrast, more advanced phonemizers with a deeper linguistic understanding typically incur high computational costs, which prevents real-time performance. This paper examines the trade-off between phonemization quality and inference speed in G2P-aided TTS systems, introducing a practical framework to bridge this gap. We propose lightweight strategies for context-aware phonemization and a service-oriented TTS architecture that executes these modules as independent services. This design decouples heavy context-aware components from the core TTS engine, effectively breaking the latency barrier and enabling real-time use of high-quality phonemization models. Experimental results confirm that the proposed system improves pronunciation soundness and linguistic accuracy while maintaining real-time responsiveness, making it well-suited for offline and end-device TTS applications.</p> - https://arxiv.org/abs/2512.08006 - Mon, 08 Dec 2025 19:49:33 +0000 + WorldPlay: Towards Long-Term Geometric Consistency for Real-Time Interactive World Modeling + https://arxiv.org/abs/2512.14614 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.14614.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Wenqiang Sun, Haiyu Zhang, Haoyuan Wang, Junta Wu, Zehan Wang, Zhenwei Wang, Yunhong Wang, Jun Zhang, Tengfei Wang, Chunchao Guo</p><p><b>Upvotes:</b> 61</p><p><b>Summary:</b> This paper presents WorldPlay, a streaming video diffusion model that enables real-time, interactive world modeling with long-term geometric consistency, resolving the trade-off between speed and memory that limits current methods. WorldPlay draws power from three key innovations. 1) We use a Dual Action Representation to enable robust action control in response to the user's keyboard and mouse inputs. 2) To enforce long-term consistency, our Reconstituted Context Memory dynamically rebuilds context from past frames and uses temporal reframing to keep geometrically important but long-past frames accessible, effectively alleviating memory attenuation. 3) We also propose Context Forcing, a novel distillation method designed for memory-aware model. Aligning memory context between the teacher and student preserves the student's capacity to use long-range information, enabling real-time speeds while preventing error drift. Taken together, WorldPlay generates long-horizon streaming 720p video at 24 FPS with superior consistency, comparing favorably with existing techniques and showing strong generalization across diverse scenes. Project page and online demo can be found: https://3d-models.hunyuan.tencent.com/world/ and https://3d.hunyuan.tencent.com/sceneTo3D.</p> + https://arxiv.org/abs/2512.14614 + Tue, 16 Dec 2025 17:22:46 +0000 - TreeGRPO: Tree-Advantage GRPO for Online RL Post-Training of Diffusion Models - https://arxiv.org/abs/2512.08153 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.08153.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zheng Ding, Weirui Ye</p><p><b>Upvotes:</b> 5</p><p><b>Summary:</b> Reinforcement learning (RL) post-training is crucial for aligning generative models with human preferences, but its prohibitive computational cost remains a major barrier to widespread adoption. We introduce TreeGRPO, a novel RL framework that dramatically improves training efficiency by recasting the denoising process as a search tree. From shared initial noise samples, TreeGRPO strategically branches to generate multiple candidate trajectories while efficiently reusing their common prefixes. This tree-structured approach delivers three key advantages: (1) High sample efficiency, achieving better performance under same training samples (2) Fine-grained credit assignment via reward backpropagation that computes step-specific advantages, overcoming the uniform credit assignment limitation of trajectory-based methods, and (3) Amortized computation where multi-child branching enables multiple policy updates per forward pass. Extensive experiments on both diffusion and flow-based models demonstrate that TreeGRPO achieves 2.4times faster training while establishing a superior Pareto frontier in the efficiency-reward trade-off space. Our method consistently outperforms GRPO baselines across multiple benchmarks and reward models, providing a scalable and effective pathway for RL-based visual generative model alignment. The project website is available at treegrpo.github.io.</p> - https://arxiv.org/abs/2512.08153 - Tue, 09 Dec 2025 01:17:34 +0000 + EVOLVE-VLA: Test-Time Training from Environment Feedback for Vision-Language-Action Models + https://arxiv.org/abs/2512.14666 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.14666.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zechen Bai, Chen Gao, Mike Zheng Shou</p><p><b>Upvotes:</b> 7</p><p><b>Summary:</b> Achieving truly adaptive embodied intelligence requires agents that learn not just by imitating static demonstrations, but by continuously improving through environmental interaction, which is akin to how humans master skills through practice. Vision-Language-Action (VLA) models have advanced robotic manipulation by leveraging large language models, yet remain fundamentally limited by Supervised Finetuning (SFT): requiring hundreds of demonstrations per task, rigidly memorizing trajectories, and failing to adapt when deployment conditions deviate from training. We introduce EVOLVE-VLA, a test-time training framework enabling VLAs to continuously adapt through environment interaction with minimal or zero task-specific demonstrations. The key technical challenge is replacing oracle reward signals (unavailable at test time) with autonomous feedback. We address this through a learned progress estimator providing dense feedback, and critically, we design our framework to ``tame'' this inherently noisy signal via two mechanisms: (1) an accumulative progress estimation mechanism smoothing noisy point-wise estimates, and (2) a progressive horizon extension strategy enabling gradual policy evolution. EVOLVE-VLA achieves substantial gains: +8.6\% on long-horizon tasks, +22.0\% in 1-shot learning, and enables cross-task generalization -- achieving 20.8\% success on unseen tasks without task-specific demonstrations training (vs. 0\% for pure SFT). Qualitative analysis reveals emergent capabilities absent in demonstrations, including error recovery and novel strategies. This work represents a critical step toward VLAs that truly learn and adapt, moving beyond static imitation toward continuous self-improvements.</p> + https://arxiv.org/abs/2512.14666 + Tue, 16 Dec 2025 18:26:38 +0000 - Ground Slow, Move Fast: A Dual-System Foundation Model for Generalizable Vision-and-Language Navigation - https://arxiv.org/abs/2512.08186 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.08186.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Meng Wei, Chenyang Wan, Jiaqi Peng, Xiqian Yu, Yuqiang Yang, Delin Feng, Wenzhe Cai, Chenming Zhu, Tai Wang, Jiangmiao Pang, Xihui Liu</p><p><b>Upvotes:</b> 20</p><p><b>Summary:</b> While recent large vision-language models (VLMs) have improved generalization in vision-language navigation (VLN), existing methods typically rely on end-to-end pipelines that map vision-language inputs directly to short-horizon discrete actions. Such designs often produce fragmented motions, incur high latency, and struggle with real-world challenges like dynamic obstacle avoidance. We propose DualVLN, the first dual-system VLN foundation model that synergistically integrates high-level reasoning with low-level action execution. System 2, a VLM-based global planner, "grounds slowly" by predicting mid-term waypoint goals via image-grounded reasoning. System 1, a lightweight, multi-modal conditioning Diffusion Transformer policy, "moves fast" by leveraging both explicit pixel goals and latent features from System 2 to generate smooth and accurate trajectories. The dual-system design enables robust real-time control and adaptive local decision-making in complex, dynamic environments. By decoupling training, the VLM retains its generalization, while System 1 achieves interpretable and effective local navigation. DualVLN outperforms prior methods across all VLN benchmarks and real-world experiments demonstrate robust long-horizon planning and real-time adaptability in dynamic environments.</p> - https://arxiv.org/abs/2512.08186 - Tue, 09 Dec 2025 02:29:36 +0000 + Fast and Accurate Causal Parallel Decoding using Jacobi Forcing + https://arxiv.org/abs/2512.14681 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.14681.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Lanxiang Hu, Siqi Kou, Yichao Fu, Samyam Rajbhandari, Tajana Rosing, Yuxiong He, Zhijie Deng, Hao Zhang</p><p><b>Upvotes:</b> 39</p><p><b>Summary:</b> Multi-token generation has emerged as a promising paradigm for accelerating transformer-based large model inference. Recent efforts primarily explore diffusion Large Language Models (dLLMs) for parallel decoding to reduce inference latency. To achieve AR-level generation quality, many techniques adapt AR models into dLLMs to enable parallel decoding. However, they suffer from limited speedup compared to AR models due to a pretrain-to-posttrain mismatch. Specifically, the masked data distribution in post-training deviates significantly from the real-world data distribution seen during pretraining, and dLLMs rely on bidirectional attention, which conflicts with the causal prior learned during pretraining and hinders the integration of exact KV cache reuse. To address this, we introduce Jacobi Forcing, a progressive distillation paradigm where models are trained on their own generated parallel decoding trajectories, smoothly shifting AR models into efficient parallel decoders while preserving their pretrained causal inference property. The models trained under this paradigm, Jacobi Forcing Model, achieves 3.8x wall-clock speedup on coding and math benchmarks with minimal loss in performance. Based on Jacobi Forcing Models' trajectory characteristics, we introduce multi-block decoding with rejection recycling, which enables up to 4.5x higher token acceptance count per iteration and nearly 4.0x wall-clock speedup, effectively trading additional compute for lower inference latency. Our code is available at https://github.com/hao-ai-lab/JacobiForcing.</p> + https://arxiv.org/abs/2512.14681 + Tue, 16 Dec 2025 18:45:18 +0000 - OpenSubject: Leveraging Video-Derived Identity and Diversity Priors for Subject-driven Image Generation and Manipulation - https://arxiv.org/abs/2512.08294 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.08294.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yexin Liu, Manyuan Zhang, Yueze Wang, Hongyu Li, Dian Zheng, Weiming Zhang, Changsheng Lu, Xunliang Cai, Yan Feng, Peng Pei, Harry Yang</p><p><b>Upvotes:</b> 17</p><p><b>Summary:</b> Despite the promising progress in subject-driven image generation, current models often deviate from the reference identities and struggle in complex scenes with multiple subjects. To address this challenge, we introduce OpenSubject, a video-derived large-scale corpus with 2.5M samples and 4.35M images for subject-driven generation and manipulation. The dataset is built with a four-stage pipeline that exploits cross-frame identity priors. (i) Video Curation. We apply resolution and aesthetic filtering to obtain high-quality clips. (ii) Cross-Frame Subject Mining and Pairing. We utilize vision-language model (VLM)-based category consensus, local grounding, and diversity-aware pairing to select image pairs. (iii) Identity-Preserving Reference Image Synthesis. We introduce segmentation map-guided outpainting to synthesize the input images for subject-driven generation and box-guided inpainting to generate input images for subject-driven manipulation, together with geometry-aware augmentations and irregular boundary erosion. (iv) Verification and Captioning. We utilize a VLM to validate synthesized samples, re-synthesize failed samples based on stage (iii), and then construct short and long captions. In addition, we introduce a benchmark covering subject-driven generation and manipulation, and then evaluate identity fidelity, prompt adherence, manipulation consistency, and background consistency with a VLM judge. Extensive experiments show that training with OpenSubject improves generation and manipulation performance, particularly in complex scenes.</p> - https://arxiv.org/abs/2512.08294 - Tue, 09 Dec 2025 06:49:33 +0000 + MMGR: Multi-Modal Generative Reasoning + https://arxiv.org/abs/2512.14691 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.14691.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zefan Cai, Haoyi Qiu, Tianyi Ma, Haozhe Zhao, Gengze Zhou, Kung-Hsiang Huang, Parisa Kordjamshidi, Minjia Zhang, Xiao Wen, Jiuxiang Gu, Nanyun Peng, Junjie Hu</p><p><b>Upvotes:</b> 112</p><p><b>Summary:</b> Video foundation models generate visually realistic and temporally coherent content, but their reliability as world simulators depends on whether they capture physical, logical, and spatial constraints. Existing metrics such as Frechet Video Distance (FVD) emphasize perceptual quality and overlook reasoning failures, including violations of causality, physics, and global consistency. We introduce MMGR (Multi-Modal Generative Reasoning Evaluation and Benchmark), a principled evaluation framework based on five reasoning abilities: Physical, Logical, 3D Spatial, 2D Spatial, and Temporal. MMGR evaluates generative reasoning across three domains: Abstract Reasoning (ARC-AGI, Sudoku), Embodied Navigation (real-world 3D navigation and localization), and Physical Commonsense (sports and compositional interactions). MMGR applies fine-grained metrics that require holistic correctness across both video and image generation. We benchmark leading video models (Veo-3, Sora-2, Wan-2.2) and image models (Nano-banana, Nano-banana Pro, GPT-4o-image, Qwen-image), revealing strong performance gaps across domains. Models show moderate success on Physical Commonsense tasks but perform poorly on Abstract Reasoning (below 10 percent accuracy on ARC-AGI) and struggle with long-horizon spatial planning in embodied settings. Our analysis highlights key limitations in current models, including overreliance on perceptual data, weak global state consistency, and objectives that reward visual plausibility over causal correctness. MMGR offers a unified diagnostic benchmark and a path toward reasoning-aware generative world models.</p> + https://arxiv.org/abs/2512.14691 + Tue, 16 Dec 2025 18:58:04 +0000 - Towards a Science of Scaling Agent Systems - https://arxiv.org/abs/2512.08296 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.08296.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yubin Kim, Ken Gu, Chanwoo Park, Chunjong Park, Samuel Schmidgall, A. Ali Heydari, Yao Yan, Zhihan Zhang, Yuchen Zhuang, Mark Malhotra, Paul Pu Liang, Hae Won Park, Yuzhe Yang, Xuhai Xu, Yilun Du, Shwetak Patel, Tim Althoff, Daniel McDuff, Xin Liu</p><p><b>Upvotes:</b> 9</p><p><b>Summary:</b> Agents, language model (LM)-based systems that are capable of reasoning, planning, and acting are becoming the dominant paradigm for real-world AI applications. Despite this widespread adoption, the principles that determine their performance remain underexplored, leaving practitioners to rely on heuristics rather than principled design choices. We address this gap by deriving quantitative scaling principles for agent systems. We evaluate this across four diverse benchmarks: Finance-Agent, BrowseComp-Plus, PlanCraft, and Workbench. Using five canonical architectures (Single, Independent, Centralized, Decentralized, Hybrid) instantiated across three LLM families, we perform a controlled evaluation spanning 180 configurations with standardized tools and token budgets. We derive a predictive model using empirical coordination metrics, including efficiency, overhead, error amplification, and redundancy, that achieves cross-validated R^2=0.513. We identify three dominant effects: (1) a tool-coordination trade-off: under fixed computational budgets, tool-heavy tasks suffer disproportionately from multi-agent overhead. (2) a capability saturation: coordination yields diminishing or negative returns (beta=-0.408, p<0.001) once single-agent baselines exceed ~45%. (3) topology-dependent error amplification: independent agents amplify errors 17.2x through unchecked propagation, while centralized coordination contains this to 4.4x. Centralized coordination improves performance by 80.9% on parallelizable tasks like financial reasoning, while decentralized coordination excels on dynamic web navigation (+9.2% vs. +0.2%). Yet for sequential reasoning tasks, all multi-agent variants degraded performance by 39-70%. The framework predicts the optimal coordination strategy for 87% of held-out configurations, providing a predictive principle of agentic scaling based on measurable task properties.</p> - https://arxiv.org/abs/2512.08296 - Tue, 09 Dec 2025 06:52:21 +0000 + Universal Reasoning Model + https://arxiv.org/abs/2512.14693 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.14693.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zitian Gao, Lynx Chen, Yihao Xiao, He Xing, Ran Tao, Haoming Luo, Joey Zhou, Bryan Dai</p><p><b>Upvotes:</b> 35</p><p><b>Summary:</b> Universal transformers (UTs) have been widely used for complex reasoning tasks such as ARC-AGI and Sudoku, yet the specific sources of their performance gains remain underexplored. In this work, we systematically analyze UTs variants and show that improvements on ARC-AGI primarily arise from the recurrent inductive bias and strong nonlinear components of Transformer, rather than from elaborate architectural designs. Motivated by this finding, we propose the Universal Reasoning Model (URM), which enhances the UT with short convolution and truncated backpropagation. Our approach substantially improves reasoning performance, achieving state-of-the-art 53.8% pass@1 on ARC-AGI 1 and 16.0% pass@1 on ARC-AGI 2. Our code is avaliable at https://github.com/zitian-gao/URM.</p> + https://arxiv.org/abs/2512.14693 + Tue, 16 Dec 2025 18:58:45 +0000 - Terrain Diffusion: A Diffusion-Based Successor to Perlin Noise in Infinite, Real-Time Terrain Generation - https://arxiv.org/abs/2512.08309 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.08309.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Alexander Goslin</p><p><b>Upvotes:</b> 2</p><p><b>Summary:</b> For decades, procedural worlds have been built on procedural noise functions such as Perlin noise, which are fast and infinite, yet fundamentally limited in realism and large-scale coherence. We introduce Terrain Diffusion, an AI-era successor to Perlin noise that bridges the fidelity of diffusion models with the properties that made procedural noise indispensable: seamless infinite extent, seed-consistency, and constant-time random access. At its core is InfiniteDiffusion, a novel algorithm for infinite generation, enabling seamless, real-time synthesis of boundless landscapes. A hierarchical stack of diffusion models couples planetary context with local detail, while a compact Laplacian encoding stabilizes outputs across Earth-scale dynamic ranges. An open-source infinite-tensor framework supports constant-memory manipulation of unbounded tensors, and few-step consistency distillation enables efficient generation. Together, these components establish diffusion models as a practical foundation for procedural world generation, capable of synthesizing entire planets coherently, controllably, and without limits.</p> - https://arxiv.org/abs/2512.08309 - Tue, 09 Dec 2025 07:10:35 +0000 + CRISP: Contact-Guided Real2Sim from Monocular Video with Planar Scene Primitives + https://arxiv.org/abs/2512.14696 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.14696.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zihan Wang, Jiashun Wang, Jeff Tan, Yiwen Zhao, Jessica Hodgins, Shubham Tulsiani, Deva Ramanan</p><p><b>Upvotes:</b> 6</p><p><b>Summary:</b> We introduce CRISP, a method that recovers simulatable human motion and scene geometry from monocular video. Prior work on joint human-scene reconstruction relies on data-driven priors and joint optimization with no physics in the loop, or recovers noisy geometry with artifacts that cause motion tracking policies with scene interactions to fail. In contrast, our key insight is to recover convex, clean, and simulation-ready geometry by fitting planar primitives to a point cloud reconstruction of the scene, via a simple clustering pipeline over depth, normals, and flow. To reconstruct scene geometry that might be occluded during interactions, we make use of human-scene contact modeling (e.g., we use human posture to reconstruct the occluded seat of a chair). Finally, we ensure that human and scene reconstructions are physically-plausible by using them to drive a humanoid controller via reinforcement learning. Our approach reduces motion tracking failure rates from 55.2\% to 6.9\% on human-centric video benchmarks (EMDB, PROX), while delivering a 43\% faster RL simulation throughput. We further validate it on in-the-wild videos including casually-captured videos, Internet videos, and even Sora-generated videos. This demonstrates CRISP's ability to generate physically-valid human motion and interaction environments at scale, greatly advancing real-to-sim applications for robotics and AR/VR.</p> + https://arxiv.org/abs/2512.14696 + Tue, 16 Dec 2025 18:59:50 +0000 - TrackingWorld: World-centric Monocular 3D Tracking of Almost All Pixels - https://arxiv.org/abs/2512.08358 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.08358.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jiahao Lu, Weitao Xiong, Jiacheng Deng, Peng Li, Tianyu Huang, Zhiyang Dou, Cheng Lin, Sai-Kit Yeung, Yuan Liu</p><p><b>Upvotes:</b> 3</p><p><b>Summary:</b> Monocular 3D tracking aims to capture the long-term motion of pixels in 3D space from a single monocular video and has witnessed rapid progress in recent years. However, we argue that the existing monocular 3D tracking methods still fall short in separating the camera motion from foreground dynamic motion and cannot densely track newly emerging dynamic subjects in the videos. To address these two limitations, we propose TrackingWorld, a novel pipeline for dense 3D tracking of almost all pixels within a world-centric 3D coordinate system. First, we introduce a tracking upsampler that efficiently lifts the arbitrary sparse 2D tracks into dense 2D tracks. Then, to generalize the current tracking methods to newly emerging objects, we apply the upsampler to all frames and reduce the redundancy of 2D tracks by eliminating the tracks in overlapped regions. Finally, we present an efficient optimization-based framework to back-project dense 2D tracks into world-centric 3D trajectories by estimating the camera poses and the 3D coordinates of these 2D tracks. Extensive evaluations on both synthetic and real-world datasets demonstrate that our system achieves accurate and dense 3D tracking in a world-centric coordinate frame.</p> - https://arxiv.org/abs/2512.08358 - Tue, 09 Dec 2025 08:35:42 +0000 + Spherical Leech Quantization for Visual Tokenization and Generation + https://arxiv.org/abs/2512.14697 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.14697.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yue Zhao, Hanwen Jiang, Zhenlin Xu, Chutong Yang, Ehsan Adeli, Philipp Krähenbühl</p><p><b>Upvotes:</b> 7</p><p><b>Summary:</b> Non-parametric quantization has received much attention due to its efficiency on parameters and scalability to a large codebook. In this paper, we present a unified formulation of different non-parametric quantization methods through the lens of lattice coding. The geometry of lattice codes explains the necessity of auxiliary loss terms when training auto-encoders with certain existing lookup-free quantization variants such as BSQ. As a step forward, we explore a few possible candidates, including random lattices, generalized Fibonacci lattices, and densest sphere packing lattices. Among all, we find the Leech lattice-based quantization method, which is dubbed as Spherical Leech Quantization (Λ_{24}-SQ), leads to both a simplified training recipe and an improved reconstruction-compression tradeoff thanks to its high symmetry and even distribution on the hypersphere. In image tokenization and compression tasks, this quantization approach achieves better reconstruction quality across all metrics than BSQ, the best prior art, while consuming slightly fewer bits. The improvement also extends to state-of-the-art auto-regressive image generation frameworks.</p> + https://arxiv.org/abs/2512.14697 + Tue, 16 Dec 2025 18:59:57 +0000 - SAM-Body4D: Training-Free 4D Human Body Mesh Recovery from Videos - https://arxiv.org/abs/2512.08406 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.08406.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Mingqi Gao, Yunqi Miao, Jungong Han</p><p><b>Upvotes:</b> 2</p><p><b>Summary:</b> Human Mesh Recovery (HMR) aims to reconstruct 3D human pose and shape from 2D observations and is fundamental to human-centric understanding in real-world scenarios. While recent image-based HMR methods such as SAM 3D Body achieve strong robustness on in-the-wild images, they rely on per-frame inference when applied to videos, leading to temporal inconsistency and degraded performance under occlusions. We address these issues without extra training by leveraging the inherent human continuity in videos. We propose SAM-Body4D, a training-free framework for temporally consistent and occlusion-robust HMR from videos. We first generate identity-consistent masklets using a promptable video segmentation model, then refine them with an Occlusion-Aware module to recover missing regions. The refined masklets guide SAM 3D Body to produce consistent full-body mesh trajectories, while a padding-based parallel strategy enables efficient multi-human inference. Experimental results demonstrate that SAM-Body4D achieves improved temporal stability and robustness in challenging in-the-wild videos, without any retraining. Our code and demo are available at: https://github.com/gaomingqi/sam-body4d.</p> - https://arxiv.org/abs/2512.08406 - Tue, 09 Dec 2025 09:37:31 +0000 + TimeLens: Rethinking Video Temporal Grounding with Multimodal LLMs + https://arxiv.org/abs/2512.14698 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.14698.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jun Zhang, Teng Wang, Yuying Ge, Yixiao Ge, Xinhao Li, Ying Shan, Limin Wang</p><p><b>Upvotes:</b> 18</p><p><b>Summary:</b> This paper does not introduce a novel method but instead establishes a straightforward, incremental, yet essential baseline for video temporal grounding (VTG), a core capability in video understanding. While multimodal large language models (MLLMs) excel at various video understanding tasks, the recipes for optimizing them for VTG remain under-explored. In this paper, we present TimeLens, a systematic investigation into building MLLMs with strong VTG ability, along two primary dimensions: data quality and algorithmic design. We first expose critical quality issues in existing VTG benchmarks and introduce TimeLens-Bench, comprising meticulously re-annotated versions of three popular benchmarks with strict quality criteria. Our analysis reveals dramatic model re-rankings compared to legacy benchmarks, confirming the unreliability of prior evaluation standards. We also address noisy training data through an automated re-annotation pipeline, yielding TimeLens-100K, a large-scale, high-quality training dataset. Building on our data foundation, we conduct in-depth explorations of algorithmic design principles, yielding a series of meaningful insights and effective yet efficient practices. These include interleaved textual encoding for time representation, a thinking-free reinforcement learning with verifiable rewards (RLVR) approach as the training paradigm, and carefully designed recipes for RLVR training. These efforts culminate in TimeLens models, a family of MLLMs with state-of-the-art VTG performance among open-source models and even surpass proprietary models such as GPT-5 and Gemini-2.5-Flash. All codes, data, and models will be released to facilitate future research.</p> + https://arxiv.org/abs/2512.14698 + Tue, 16 Dec 2025 18:59:58 +0000 - Visionary: The World Model Carrier Built on WebGPU-Powered Gaussian Splatting Platform - https://arxiv.org/abs/2512.08478 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.08478.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yuning Gong, Yifei Liu, Yifan Zhan, Muyao Niu, Xueying Li, Yuanjun Liao, Jiaming Chen, Yuanyuan Gao, Jiaqi Chen, Minming Chen, Li Zhou, Yuning Zhang, Wei Wang, Xiaoqing Hou, Huaxi Huang, Shixiang Tang, Le Ma, Dingwen Zhang, Xue Yang, Junchi Yan, Yanchi Zhang, Yinqiang Zheng, Xiao Sun, Zhihang Zhong</p><p><b>Upvotes:</b> 73</p><p><b>Summary:</b> Neural rendering, particularly 3D Gaussian Splatting (3DGS), has evolved rapidly and become a key component for building world models. However, existing viewer solutions remain fragmented, heavy, or constrained by legacy pipelines, resulting in high deployment friction and limited support for dynamic content and generative models. In this work, we present Visionary, an open, web-native platform for real-time various Gaussian Splatting and meshes rendering. Built on an efficient WebGPU renderer with per-frame ONNX inference, Visionary enables dynamic neural processing while maintaining a lightweight, "click-to-run" browser experience. It introduces a standardized Gaussian Generator contract, which not only supports standard 3DGS rendering but also allows plug-and-play algorithms to generate or update Gaussians each frame. Such inference also enables us to apply feedforward generative post-processing. The platform further offers a plug in three.js library with a concise TypeScript API for seamless integration into existing web applications. Experiments show that, under identical 3DGS assets, Visionary achieves superior rendering efficiency compared to current Web viewers due to GPU-based primitive sorting. It already supports multiple variants, including MLP-based 3DGS, 4DGS, neural avatars, and style transformation or enhancement networks. By unifying inference and rendering directly in the browser, Visionary significantly lowers the barrier to reproduction, comparison, and deployment of 3DGS-family methods, serving as a unified World Model Carrier for both reconstructive and generative paradigms.</p> - https://arxiv.org/abs/2512.08478 - Tue, 09 Dec 2025 10:54:58 +0000 + MemFlow: Flowing Adaptive Memory for Consistent and Efficient Long Video Narratives + https://arxiv.org/abs/2512.14699 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.14699.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Sihui Ji, Xi Chen, Shuai Yang, Xin Tao, Pengfei Wan, Hengshuang Zhao</p><p><b>Upvotes:</b> 26</p><p><b>Summary:</b> The core challenge for streaming video generation is maintaining the content consistency in long context, which poses high requirement for the memory design. Most existing solutions maintain the memory by compressing historical frames with predefined strategies. However, different to-generate video chunks should refer to different historical cues, which is hard to satisfy with fixed strategies. In this work, we propose MemFlow to address this problem. Specifically, before generating the coming chunk, we dynamically update the memory bank by retrieving the most relevant historical frames with the text prompt of this chunk. This design enables narrative coherence even if new event happens or scenario switches in future frames. In addition, during generation, we only activate the most relevant tokens in the memory bank for each query in the attention layers, which effectively guarantees the generation efficiency. In this way, MemFlow achieves outstanding long-context consistency with negligible computation burden (7.9% speed reduction compared with the memory-free baseline) and keeps the compatibility with any streaming video generation model with KV cache.</p> + https://arxiv.org/abs/2512.14699 + Tue, 16 Dec 2025 18:59:59 +0000 - Thinking with Images via Self-Calling Agent - https://arxiv.org/abs/2512.08511 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.08511.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Wenxi Yang, Yuzhong Zhao, Fang Wan, Qixiang Ye</p><p><b>Upvotes:</b> 17</p><p><b>Summary:</b> Thinking-with-images paradigms have showcased remarkable visual reasoning capability by integrating visual information as dynamic elements into the Chain-of-Thought (CoT). However, optimizing interleaved multimodal CoT (iMCoT) through reinforcement learning remains challenging, as it relies on scarce high-quality reasoning data. In this study, we propose Self-Calling Chain-of-Thought (sCoT), a novel visual reasoning paradigm that reformulates iMCoT as a language-only CoT with self-calling. Specifically, a main agent decomposes the complex visual reasoning task to atomic subtasks and invokes its virtual replicas, i.e. parameter-sharing subagents, to solve them in isolated context. sCoT enjoys substantial training effectiveness and efficiency, as it requires no explicit interleaving between modalities. sCoT employs group-relative policy optimization to reinforce effective reasoning behavior to enhance optimization. Experiments on HR-Bench 4K show that sCoT improves the overall reasoning performance by up to 1.9% with sim 75% fewer GPU hours compared to strong baseline approaches. Code is available at https://github.com/YWenxi/think-with-images-through-self-calling.</p> - https://arxiv.org/abs/2512.08511 - Tue, 09 Dec 2025 11:53:21 +0000 + Puzzle Curriculum GRPO for Vision-Centric Reasoning + https://arxiv.org/abs/2512.14944 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.14944.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Ahmadreza Jeddi, Hakki Can Karaimer, Hue Nguyen, Zhongling Wang, Ke Zhao, Javad Rajabi, Ran Zhang, Raghav Goyal, Babak Taati, Radek Grzeszczuk</p><p><b>Upvotes:</b> 33</p><p><b>Summary:</b> Recent reinforcement learning (RL) approaches like outcome-supervised GRPO have advanced chain-of-thought reasoning in Vision Language Models (VLMs), yet key issues linger: (i) reliance on costly and noisy hand-curated annotations or external verifiers; (ii) flat and sparse reward schemes in GRPO; and (iii) logical inconsistency between a chain's reasoning and its final answer. We present Puzzle Curriculum GRPO (PC-GRPO), a supervision-free recipe for RL with Verifiable Rewards (RLVR) that strengthens visual reasoning in VLMs without annotations or external verifiers. PC-GRPO replaces labels with three self-supervised puzzle environments: PatchFit, Rotation (with binary rewards) and Jigsaw (with graded partial credit mitigating reward sparsity). To counter flat rewards and vanishing group-relative advantages, we introduce a difficulty-aware curriculum that dynamically weights samples and peaks at medium difficulty. We further monitor Reasoning-Answer Consistency (RAC) during post-training: mirroring reports for vanilla GRPO in LLMs, RAC typically rises early then degrades; our curriculum delays this decline, and consistency-enforcing reward schemes further boost RAC. RAC correlates with downstream accuracy. Across diverse benchmarks and on Qwen-7B and Qwen-3B backbones, PC-GRPO improves reasoning quality, training stability, and end-task accuracy, offering a practical path to scalable, verifiable, and interpretable RL post-training for VLMs.</p> + https://arxiv.org/abs/2512.14944 + Tue, 16 Dec 2025 22:17:25 +0000 - BrainExplore: Large-Scale Discovery of Interpretable Visual Representations in the Human Brain - https://arxiv.org/abs/2512.08560 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.08560.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Navve Wasserman, Matias Cosarinsky, Yuval Golbari, Aude Oliva, Antonio Torralba, Tamar Rott Shaham, Michal Irani</p><p><b>Upvotes:</b> 35</p><p><b>Summary:</b> Understanding how the human brain represents visual concepts, and in which brain regions these representations are encoded, remains a long-standing challenge. Decades of work have advanced our understanding of visual representations, yet brain signals remain large and complex, and the space of possible visual concepts is vast. As a result, most studies remain small-scale, rely on manual inspection, focus on specific regions and properties, and rarely include systematic validation. We present a large-scale, automated framework for discovering and explaining visual representations across the human cortex. Our method comprises two main stages. First, we discover candidate interpretable patterns in fMRI activity through unsupervised, data-driven decomposition methods. Next, we explain each pattern by identifying the set of natural images that most strongly elicit it and generating a natural-language description of their shared visual meaning. To scale this process, we introduce an automated pipeline that tests multiple candidate explanations, assigns quantitative reliability scores, and selects the most consistent description for each voxel pattern. Our framework reveals thousands of interpretable patterns spanning many distinct visual concepts, including fine-grained representations previously unreported.</p> - https://arxiv.org/abs/2512.08560 - Tue, 09 Dec 2025 13:01:17 +0000 + Is Nano Banana Pro a Low-Level Vision All-Rounder? A Comprehensive Evaluation on 14 Tasks and 40 Datasets + https://arxiv.org/abs/2512.15110 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.15110.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jialong Zuo, Haoyou Deng, Hanyu Zhou, Jiaxin Zhu, Yicheng Zhang, Yiwei Zhang, Yongxin Yan, Kaixing Huang, Weisen Chen, Yongtai Deng, Rui Jin, Nong Sang, Changxin Gao</p><p><b>Upvotes:</b> 6</p><p><b>Summary:</b> The rapid evolution of text-to-image generation models has revolutionized visual content creation. While commercial products like Nano Banana Pro have garnered significant attention, their potential as generalist solvers for traditional low-level vision challenges remains largely underexplored. In this study, we investigate the critical question: Is Nano Banana Pro a Low-Level Vision All-Rounder? We conducted a comprehensive zero-shot evaluation across 14 distinct low-level tasks spanning 40 diverse datasets. By utilizing simple textual prompts without fine-tuning, we benchmarked Nano Banana Pro against state-of-the-art specialist models. Our extensive analysis reveals a distinct performance dichotomy: while Nano Banana Pro demonstrates superior subjective visual quality, often hallucinating plausible high-frequency details that surpass specialist models, it lags behind in traditional reference-based quantitative metrics. We attribute this discrepancy to the inherent stochasticity of generative models, which struggle to maintain the strict pixel-level consistency required by conventional metrics. This report identifies Nano Banana Pro as a capable zero-shot contender for low-level vision tasks, while highlighting that achieving the high fidelity of domain specialists remains a significant hurdle.</p> + https://arxiv.org/abs/2512.15110 + Wed, 17 Dec 2025 06:02:25 +0000 - Modular Neural Image Signal Processing - https://arxiv.org/abs/2512.08564 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.08564.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Mahmoud Afifi, Zhongling Wang, Ran Zhang, Michael S. Brown</p><p><b>Upvotes:</b> 5</p><p><b>Summary:</b> This paper presents a modular neural image signal processing (ISP) framework that processes raw inputs and renders high-quality display-referred images. Unlike prior neural ISP designs, our method introduces a high degree of modularity, providing full control over multiple intermediate stages of the rendering process.~This modular design not only achieves high rendering accuracy but also improves scalability, debuggability, generalization to unseen cameras, and flexibility to match different user-preference styles. To demonstrate the advantages of this design, we built a user-interactive photo-editing tool that leverages our neural ISP to support diverse editing operations and picture styles. The tool is carefully engineered to take advantage of the high-quality rendering of our neural ISP and to enable unlimited post-editable re-rendering. Our method is a fully learning-based framework with variants of different capacities, all of moderate size (ranging from ~0.5 M to ~3.9 M parameters for the entire pipeline), and consistently delivers competitive qualitative and quantitative results across multiple test sets. Watch the supplemental video at: https://youtu.be/ByhQjQSjxVM</p> - https://arxiv.org/abs/2512.08564 - Tue, 09 Dec 2025 13:04:08 +0000 + DEER: Draft with Diffusion, Verify with Autoregressive Models + https://arxiv.org/abs/2512.15176 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.15176.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zicong Cheng, Guo-Wei Yang, Jia Li, Zhijie Deng, Meng-Hao Guo, Shi-Min Hu</p><p><b>Upvotes:</b> 40</p><p><b>Summary:</b> Efficiency, as a critical practical challenge for LLM-driven agentic and reasoning systems, is increasingly constrained by the inherent latency of autoregressive (AR) decoding. Speculative decoding mitigates this cost through a draft-verify scheme, yet existing approaches rely on AR draft models (a.k.a., drafters), which introduce two fundamental issues: (1) step-wise uncertainty accumulation leads to a progressive collapse of trust between the target model and the drafter, and (2) inherently sequential decoding of AR drafters. Together, these factors cause limited speedups. In this paper, we show that a diffusion large language model (dLLM) drafters can naturally overcome these issues through its fundamentally different probabilistic modeling and efficient parallel decoding strategy. Building on this insight, we introduce DEER, an efficient speculative decoding framework that drafts with diffusion and verifies with AR models. To enable high-quality drafting, DEER employs a two-stage training pipeline to align the dLLM-based drafters with the target AR model, and further adopts single-step decoding to generate long draft segments. Experiments show DEER reaches draft acceptance lengths of up to 32 tokens, far surpassing the 10 tokens achieved by EAGLE-3. Moreover, on HumanEval with Qwen3-30B-A3B, DEER attains a 5.54x speedup, while EAGLE-3 achieves only 2.41x. Code, model, demo, etc, will be available at https://czc726.github.io/DEER/</p> + https://arxiv.org/abs/2512.15176 + Wed, 17 Dec 2025 08:19:04 +0000 - Wan-Move: Motion-controllable Video Generation via Latent Trajectory Guidance - https://arxiv.org/abs/2512.08765 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.08765.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Ruihang Chu, Yefei He, Zhekai Chen, Shiwei Zhang, Xiaogang Xu, Bin Xia, Dingdong Wang, Hongwei Yi, Xihui Liu, Hengshuang Zhao, Yu Liu, Yingya Zhang, Yujiu Yang</p><p><b>Upvotes:</b> 121</p><p><b>Summary:</b> We present Wan-Move, a simple and scalable framework that brings motion control to video generative models. Existing motion-controllable methods typically suffer from coarse control granularity and limited scalability, leaving their outputs insufficient for practical use. We narrow this gap by achieving precise and high-quality motion control. Our core idea is to directly make the original condition features motion-aware for guiding video synthesis. To this end, we first represent object motions with dense point trajectories, allowing fine-grained control over the scene. We then project these trajectories into latent space and propagate the first frame's features along each trajectory, producing an aligned spatiotemporal feature map that tells how each scene element should move. This feature map serves as the updated latent condition, which is naturally integrated into the off-the-shelf image-to-video model, e.g., Wan-I2V-14B, as motion guidance without any architecture change. It removes the need for auxiliary motion encoders and makes fine-tuning base models easily scalable. Through scaled training, Wan-Move generates 5-second, 480p videos whose motion controllability rivals Kling 1.5 Pro's commercial Motion Brush, as indicated by user studies. To support comprehensive evaluation, we further design MoveBench, a rigorously curated benchmark featuring diverse content categories and hybrid-verified annotations. It is distinguished by larger data volume, longer video durations, and high-quality motion annotations. Extensive experiments on MoveBench and the public dataset consistently show Wan-Move's superior motion quality. Code, models, and benchmark data are made publicly available.</p> - https://arxiv.org/abs/2512.08765 - Tue, 09 Dec 2025 16:13:55 +0000 + Robust and Calibrated Detection of Authentic Multimedia Content + https://arxiv.org/abs/2512.15182 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.15182.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Sarim Hashmi, Abdelrahman Elsayed, Mohammed Talha Alam, Samuele Poppi, Nils Lukas</p><p><b>Upvotes:</b> 15</p><p><b>Summary:</b> Generative models can synthesize highly realistic content, so-called deepfakes, that are already being misused at scale to undermine digital media authenticity. Current deepfake detection methods are unreliable for two reasons: (i) distinguishing inauthentic content post-hoc is often impossible (e.g., with memorized samples), leading to an unbounded false positive rate (FPR); and (ii) detection lacks robustness, as adversaries can adapt to known detectors with near-perfect accuracy using minimal computational resources. To address these limitations, we propose a resynthesis framework to determine if a sample is authentic or if its authenticity can be plausibly denied. We make two key contributions focusing on the high-precision, low-recall setting against efficient (i.e., compute-restricted) adversaries. First, we demonstrate that our calibrated resynthesis method is the most reliable approach for verifying authentic samples while maintaining controllable, low FPRs. Second, we show that our method achieves adversarial robustness against efficient adversaries, whereas prior methods are easily evaded under identical compute budgets. Our approach supports multiple modalities and leverages state-of-the-art inversion techniques.</p> + https://arxiv.org/abs/2512.15182 + Wed, 17 Dec 2025 08:31:40 +0000 - InfiniteVL: Synergizing Linear and Sparse Attention for Highly-Efficient, Unlimited-Input Vision-Language Models - https://arxiv.org/abs/2512.08829 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.08829.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Hongyuan Tao, Bencheng Liao, Shaoyu Chen, Haoran Yin, Qian Zhang, Wenyu Liu, Xinggang Wang</p><p><b>Upvotes:</b> 15</p><p><b>Summary:</b> Window attention and linear attention represent two principal strategies for mitigating the quadratic complexity and ever-growing KV cache in Vision-Language Models (VLMs). However, we observe that window-based VLMs suffer performance degradation when sequence length exceeds the window size, while linear attention underperforms on information-intensive tasks such as OCR and document understanding. To overcome these limitations, we propose InfiniteVL, a linear-complexity VLM architecture that synergizes sliding window attention (SWA) with Gated DeltaNet. For achieving competitive multimodal performance under constrained resources, we design a three-stage training strategy comprising distillation pretraining, instruction tuning, and long-sequence SFT. Remarkably, using less than 2\% of the training data required by leading VLMs, InfiniteVL not only substantially outperforms previous linear-complexity VLMs but also matches the performance of leading Transformer-based VLMs, while demonstrating effective long-term memory retention. Compared to similar-sized Transformer-based VLMs accelerated by FlashAttention-2, InfiniteVL achieves over 3.6\times inference speedup while maintaining constant latency and memory footprint. In streaming video understanding scenarios, it sustains a stable 24 FPS real-time prefill speed while preserving long-term memory cache. Code and models are available at https://github.com/hustvl/InfiniteVL.</p> - https://arxiv.org/abs/2512.08829 - Tue, 09 Dec 2025 17:18:32 +0000 + Step-GUI Technical Report + https://arxiv.org/abs/2512.15431 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.15431.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Haolong Yan, Jia Wang, Xin Huang, Yeqing Shen, Ziyang Meng, Zhimin Fan, Kaijun Tan, Jin Gao, Lieyu Shi, Mi Yang, Shiliang Yang, Zhirui Wang, Brian Li, Kang An, Chenyang Li, Lei Lei, Mengmeng Duan, Danxun Liang, Guodong Liu, Hang Cheng, Hao Wu, Jie Dong, Junhao Huang, Mei Chen, Renjie Yu, Shunshan Li, Xu Zhou, Yiting Dai, Yineng Deng, Yingdan Liang, Zelin Chen, Wen Sun, Chengxu Yan, Chunqin Xu, Dong Li, Fengqiong Xiao, Guanghao Fan, Guopeng Li, Guozhen Peng, Hongbing Li, Hang Li, Hongming Chen, Jingjing Xie, Jianyong Li, Jingyang Zhang, Jiaju Ren, Jiayu Yuan, Jianpeng Yin, Kai Cao, Liang Zhao, Liguo Tan, Liying Shi, Mengqiang Ren, Min Xu, Manjiao Liu, Mao Luo, Mingxin Wan, Na Wang, Nan Wu, Ning Wang, Peiyao Ma, Qingzhou Zhang, Qiao Wang, Qinlin Zeng, Qiong Gao, Qiongyao Li, Shangwu Zhong, Shuli Gao, Shaofan Liu, Shisi Gao, Shuang Luo, Xingbin Liu, Xiaojia Liu, Xiaojie Hou, Xin Liu, Xuanti Feng, Xuedan Cai, Xuan Wen, Xianwei Zhu, Xin Liang, Xin Liu, Xin Zhou, Yingxiu Zhao, Yukang Shi, Yunfang Xu, Yuqing Zeng, Yixun Zhang, Zejia Weng, Zhonghao Yan, Zhiguo Huang, Zhuoyu Wang, Zheng Ge, Jing Li, Yibo Zhu, Binxing Jiao, Xiangyu Zhang, Daxin Jiang</p><p><b>Upvotes:</b> 118</p><p><b>Summary:</b> Recent advances in multimodal large language models unlock unprecedented opportunities for GUI automation. However, a fundamental challenge remains: how to efficiently acquire high-quality training data while maintaining annotation reliability? We introduce a self-evolving training pipeline powered by the Calibrated Step Reward System, which converts model-generated trajectories into reliable training signals through trajectory-level calibration, achieving >90% annotation accuracy with 10-100x lower cost. Leveraging this pipeline, we introduce Step-GUI, a family of models (4B/8B) that achieves state-of-the-art GUI performance (8B: 80.2% AndroidWorld, 48.5% OSWorld, 62.6% ScreenShot-Pro) while maintaining robust general capabilities. As GUI agent capabilities improve, practical deployment demands standardized interfaces across heterogeneous devices while protecting user privacy. To this end, we propose GUI-MCP, the first Model Context Protocol for GUI automation with hierarchical architecture that combines low-level atomic operations and high-level task delegation to local specialist models, enabling high-privacy execution where sensitive data stays on-device. Finally, to assess whether agents can handle authentic everyday usage, we introduce AndroidDaily, a benchmark grounded in real-world mobile usage patterns with 3146 static actions and 235 end-to-end tasks across high-frequency daily scenarios (8B: static 89.91%, end-to-end 52.50%). Our work advances the development of practical GUI agents and demonstrates strong potential for real-world deployment in everyday digital interactions.</p> + https://arxiv.org/abs/2512.15431 + Wed, 17 Dec 2025 13:26:30 +0000 - Fed-SE: Federated Self-Evolution for Privacy-Constrained Multi-Environment LLM Agents - https://arxiv.org/abs/2512.08870 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.08870.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Xiang Chen, Yuling Shi, Qizhen Lan, Yuchao Qiu, Xiaodong Gu</p><p><b>Upvotes:</b> 3</p><p><b>Summary:</b> LLM agents are widely deployed in complex interactive tasks, yet privacy constraints often preclude centralized optimization and co-evolution across dynamic environments. While Federated Learning (FL) has proven effective on static datasets, its extension to the open-ended self-evolution of agents remains underexplored. Directly applying standard FL is challenging: heterogeneous tasks and sparse, trajectory-level rewards introduce severe gradient conflicts, destabilizing the global optimization process. To bridge this gap, we propose Fed-SE, a Federated Self-Evolution framework for LLM agents. Fed-SE establishes a local evolution-global aggregation paradigm. Locally, agents employ parameter-efficient fine-tuning on filtered, high-return trajectories to achieve stable gradient updates. Globally, Fed-SE aggregates updates within a low-rank subspace that disentangles environment-specific dynamics, effectively reducing negative transfer across clients. Experiments across five heterogeneous environments demonstrate that Fed-SE improves average task success rates by approximately 18% over federated baselines, validating its effectiveness in robust cross-environment knowledge transfer in privacy-constrained deployments.</p> - https://arxiv.org/abs/2512.08870 - Tue, 09 Dec 2025 18:04:41 +0000 + Qwen-Image-Layered: Towards Inherent Editability via Layer Decomposition + https://arxiv.org/abs/2512.15603 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.15603.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Shengming Yin, Zekai Zhang, Zecheng Tang, Kaiyuan Gao, Xiao Xu, Kun Yan, Jiahao Li, Yilei Chen, Yuxiang Chen, Heung-Yeung Shum, Lionel M. Ni, Jingren Zhou, Junyang Lin, Chenfei Wu</p><p><b>Upvotes:</b> 41</p><p><b>Summary:</b> Recent visual generative models often struggle with consistency during image editing due to the entangled nature of raster images, where all visual content is fused into a single canvas. In contrast, professional design tools employ layered representations, allowing isolated edits while preserving consistency. Motivated by this, we propose Qwen-Image-Layered, an end-to-end diffusion model that decomposes a single RGB image into multiple semantically disentangled RGBA layers, enabling inherent editability, where each RGBA layer can be independently manipulated without affecting other content. To support variable-length decomposition, we introduce three key components: (1) an RGBA-VAE to unify the latent representations of RGB and RGBA images; (2) a VLD-MMDiT (Variable Layers Decomposition MMDiT) architecture capable of decomposing a variable number of image layers; and (3) a Multi-stage Training strategy to adapt a pretrained image generation model into a multilayer image decomposer. Furthermore, to address the scarcity of high-quality multilayer training images, we build a pipeline to extract and annotate multilayer images from Photoshop documents (PSD). Experiments demonstrate that our method significantly surpasses existing approaches in decomposition quality and establishes a new paradigm for consistent image editing. Our code and models are released on https://github.com/QwenLM/Qwen-Image-Layered{https://github.com/QwenLM/Qwen-Image-Layered}</p> + https://arxiv.org/abs/2512.15603 + Wed, 17 Dec 2025 17:12:42 +0000 - Efficiently Reconstructing Dynamic Scenes One D4RT at a Time - https://arxiv.org/abs/2512.08924 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.08924.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Chuhan Zhang, Guillaume Le Moing, Skanda Koppula, Ignacio Rocco, Liliane Momeni, Junyu Xie, Shuyang Sun, Rahul Sukthankar, Joëlle K Barral, Raia Hadsell, Zoubin Ghahramani, Andrew Zisserman, Junlin Zhang, Mehdi SM Sajjadi</p><p><b>Upvotes:</b> 4</p><p><b>Summary:</b> Understanding and reconstructing the complex geometry and motion of dynamic scenes from video remains a formidable challenge in computer vision. This paper introduces D4RT, a simple yet powerful feedforward model designed to efficiently solve this task. D4RT utilizes a unified transformer architecture to jointly infer depth, spatio-temporal correspondence, and full camera parameters from a single video. Its core innovation is a novel querying mechanism that sidesteps the heavy computation of dense, per-frame decoding and the complexity of managing multiple, task-specific decoders. Our decoding interface allows the model to independently and flexibly probe the 3D position of any point in space and time. The result is a lightweight and highly scalable method that enables remarkably efficient training and inference. We demonstrate that our approach sets a new state of the art, outperforming previous methods across a wide spectrum of 4D reconstruction tasks. We refer to the project webpage for animated results: https://d4rt-paper.github.io/.</p> - https://arxiv.org/abs/2512.08924 - Tue, 09 Dec 2025 18:57:21 +0000 + IC-Effect: Precise and Efficient Video Effects Editing via In-Context Learning + https://arxiv.org/abs/2512.15635 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.15635.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yuanhang Li, Yiren Song, Junzhe Bai, Xinran Liang, Hu Yang, Libiao Jin, Qi Mao</p><p><b>Upvotes:</b> 19</p><p><b>Summary:</b> We propose IC-Effect, an instruction-guided, DiT-based framework for few-shot video VFX editing that synthesizes complex effects (\eg flames, particles and cartoon characters) while strictly preserving spatial and temporal consistency. Video VFX editing is highly challenging because injected effects must blend seamlessly with the background, the background must remain entirely unchanged, and effect patterns must be learned efficiently from limited paired data. However, existing video editing models fail to satisfy these requirements. IC-Effect leverages the source video as clean contextual conditions, exploiting the contextual learning capability of DiT models to achieve precise background preservation and natural effect injection. A two-stage training strategy, consisting of general editing adaptation followed by effect-specific learning via Effect-LoRA, ensures strong instruction following and robust effect modeling. To further improve efficiency, we introduce spatiotemporal sparse tokenization, enabling high fidelity with substantially reduced computation. We also release a paired VFX editing dataset spanning 15 high-quality visual styles. Extensive experiments show that IC-Effect delivers high-quality, controllable, and temporally consistent VFX editing, opening new possibilities for video creation.</p> + https://arxiv.org/abs/2512.15635 + Wed, 17 Dec 2025 17:47:18 +0000 - Learning Unmasking Policies for Diffusion Language Models - https://arxiv.org/abs/2512.09106 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.09106.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Metod Jazbec, Theo X. Olausson, Louis Béthune, Pierre Ablin, Michael Kirchhof, Joao Monterio, Victor Turrisi, Jason Ramapuram, Marco Cuturi</p><p><b>Upvotes:</b> 6</p><p><b>Summary:</b> Diffusion (Large) Language Models (dLLMs) now match the downstream performance of their autoregressive counterparts on many tasks, while holding the promise of being more efficient during inference. One particularly successful variant is masked discrete diffusion, in which a buffer filled with special mask tokens is progressively replaced with tokens sampled from the model's vocabulary. Efficiency can be gained by unmasking several tokens in parallel, but doing too many at once risks degrading the generation quality. Thus, one critical design aspect of dLLMs is the sampling procedure that selects, at each step of the diffusion process, which tokens to replace. Indeed, recent work has found that heuristic strategies such as confidence thresholding lead to both higher quality and token throughput compared to random unmasking. However, such heuristics have downsides: they require manual tuning, and we observe that their performance degrades with larger buffer sizes. In this work, we instead propose to train sampling procedures using reinforcement learning. Specifically, we formalize masked diffusion sampling as a Markov decision process in which the dLLM serves as the environment, and propose a lightweight policy architecture based on a single-layer transformer that maps dLLM token confidences to unmasking decisions. Our experiments show that these trained policies match the performance of state-of-the-art heuristics when combined with semi-autoregressive generation, while outperforming them in the full diffusion setting. We also examine the transferability of these policies, finding that they can generalize to new underlying dLLMs and longer sequence lengths. However, we also observe that their performance degrades when applied to out-of-domain data, and that fine-grained tuning of the accuracy-efficiency trade-off can be challenging with our approach.</p> - https://arxiv.org/abs/2512.09106 - Tue, 09 Dec 2025 20:44:33 +0000 + VTCBench: Can Vision-Language Models Understand Long Context with Vision-Text Compression? + https://arxiv.org/abs/2512.15649 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.15649.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Hongbo Zhao, Meng Wang, Fei Zhu, Wenzhuo Liu, Bolin Ni, Fanhu Zeng, Gaofeng Meng, Zhaoxiang Zhang</p><p><b>Upvotes:</b> 6</p><p><b>Summary:</b> The computational and memory overheads associated with expanding the context window of LLMs severely limit their scalability. A noteworthy solution is vision-text compression (VTC), exemplified by frameworks like DeepSeek-OCR and Glyph, which convert long texts into dense 2D visual representations, thereby achieving token compression ratios of 3x-20x. However, the impact of this high information density on the core long-context capabilities of vision-language models (VLMs) remains under-investigated. To address this gap, we introduce the first benchmark for VTC and systematically assess the performance of VLMs across three long-context understanding settings: VTC-Retrieval, which evaluates the model's ability to retrieve and aggregate information; VTC-Reasoning, which requires models to infer latent associations to locate facts with minimal lexical overlap; and VTC-Memory, which measures comprehensive question answering within long-term dialogue memory. Furthermore, we establish the VTCBench-Wild to simulate diverse input scenarios.We comprehensively evaluate leading open-source and proprietary models on our benchmarks. The results indicate that, despite being able to decode textual information (e.g., OCR) well, most VLMs exhibit a surprisingly poor long-context understanding ability with VTC-compressed information, failing to capture long associations or dependencies in the context.This study provides a deep understanding of VTC and serves as a foundation for designing more efficient and scalable VLMs.</p> + https://arxiv.org/abs/2512.15649 + Wed, 17 Dec 2025 17:58:35 +0000 - WonderZoom: Multi-Scale 3D World Generation - https://arxiv.org/abs/2512.09164 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.09164.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jin Cao, Hong-Xing Yu, Jiajun Wu</p><p><b>Upvotes:</b> 11</p><p><b>Summary:</b> We present WonderZoom, a novel approach to generating 3D scenes with contents across multiple spatial scales from a single image. Existing 3D world generation models remain limited to single-scale synthesis and cannot produce coherent scene contents at varying granularities. The fundamental challenge is the lack of a scale-aware 3D representation capable of generating and rendering content with largely different spatial sizes. WonderZoom addresses this through two key innovations: (1) scale-adaptive Gaussian surfels for generating and real-time rendering of multi-scale 3D scenes, and (2) a progressive detail synthesizer that iteratively generates finer-scale 3D contents. Our approach enables users to "zoom into" a 3D region and auto-regressively synthesize previously non-existent fine details from landscapes to microscopic features. Experiments demonstrate that WonderZoom significantly outperforms state-of-the-art video and 3D models in both quality and alignment, enabling multi-scale 3D world creation from a single image. We show video results and an interactive viewer of generated multi-scale 3D worlds in https://wonderzoom.github.io/</p> - https://arxiv.org/abs/2512.09164 - Tue, 09 Dec 2025 22:21:07 +0000 + Can LLMs Guide Their Own Exploration? Gradient-Guided Reinforcement Learning for LLM Reasoning + https://arxiv.org/abs/2512.15687 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.15687.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zhenwen Liang, Sidi Lu, Wenhao Yu, Kishan Panaganti, Yujun Zhou, Haitao Mi, Dong Yu</p><p><b>Upvotes:</b> 16</p><p><b>Summary:</b> Reinforcement learning has become essential for strengthening the reasoning abilities of large language models, yet current exploration mechanisms remain fundamentally misaligned with how these models actually learn. Entropy bonuses and external semantic comparators encourage surface level variation but offer no guarantee that sampled trajectories differ in the update directions that shape optimization. We propose G2RL, a gradient guided reinforcement learning framework in which exploration is driven not by external heuristics but by the model own first order update geometry. For each response, G2RL constructs a sequence level feature from the model final layer sensitivity, obtainable at negligible cost from a standard forward pass, and measures how each trajectory would reshape the policy by comparing these features within a sampled group. Trajectories that introduce novel gradient directions receive a bounded multiplicative reward scaler, while redundant or off manifold updates are deemphasized, yielding a self referential exploration signal that is naturally aligned with PPO style stability and KL control. Across math and general reasoning benchmarks (MATH500, AMC, AIME24, AIME25, GPQA, MMLUpro) on Qwen3 base 1.7B and 4B models, G2RL consistently improves pass@1, maj@16, and pass@k over entropy based GRPO and external embedding methods. Analyzing the induced geometry, we find that G2RL expands exploration into substantially more orthogonal and often opposing gradient directions while maintaining semantic coherence, revealing that a policy own update space provides a far more faithful and effective basis for guiding exploration in large language model reinforcement learning.</p> + https://arxiv.org/abs/2512.15687 + Wed, 17 Dec 2025 18:44:45 +0000 - OmniPSD: Layered PSD Generation with Diffusion Transformer - https://arxiv.org/abs/2512.09247 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.09247.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Cheng Liu, Yiren Song, Haofan Wang, Mike Zheng Shou</p><p><b>Upvotes:</b> 42</p><p><b>Summary:</b> Recent advances in diffusion models have greatly improved image generation and editing, yet generating or reconstructing layered PSD files with transparent alpha channels remains highly challenging. We propose OmniPSD, a unified diffusion framework built upon the Flux ecosystem that enables both text-to-PSD generation and image-to-PSD decomposition through in-context learning. For text-to-PSD generation, OmniPSD arranges multiple target layers spatially into a single canvas and learns their compositional relationships through spatial attention, producing semantically coherent and hierarchically structured layers. For image-to-PSD decomposition, it performs iterative in-context editing, progressively extracting and erasing textual and foreground components to reconstruct editable PSD layers from a single flattened image. An RGBA-VAE is employed as an auxiliary representation module to preserve transparency without affecting structure learning. Extensive experiments on our new RGBA-layered dataset demonstrate that OmniPSD achieves high-fidelity generation, structural consistency, and transparency awareness, offering a new paradigm for layered design generation and decomposition with diffusion transformers.</p> - https://arxiv.org/abs/2512.09247 - Wed, 10 Dec 2025 02:09:59 +0000 + Skyra: AI-Generated Video Detection via Grounded Artifact Reasoning + https://arxiv.org/abs/2512.15693 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.15693.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yifei Li, Wenzhao Zheng, Yanran Zhang, Runze Sun, Yu Zheng, Lei Chen, Jie Zhou, Jiwen Lu</p><p><b>Upvotes:</b> 16</p><p><b>Summary:</b> The misuse of AI-driven video generation technologies has raised serious social concerns, highlighting the urgent need for reliable AI-generated video detectors. However, most existing methods are limited to binary classification and lack the necessary explanations for human interpretation. In this paper, we present Skyra, a specialized multimodal large language model (MLLM) that identifies human-perceivable visual artifacts in AI-generated videos and leverages them as grounded evidence for both detection and explanation. To support this objective, we construct ViF-CoT-4K for Supervised Fine-Tuning (SFT), which represents the first large-scale AI-generated video artifact dataset with fine-grained human annotations. We then develop a two-stage training strategy that systematically enhances our model's spatio-temporal artifact perception, explanation capability, and detection accuracy. To comprehensively evaluate Skyra, we introduce ViF-Bench, a benchmark comprising 3K high-quality samples generated by over ten state-of-the-art video generators. Extensive experiments demonstrate that Skyra surpasses existing methods across multiple benchmarks, while our evaluation yields valuable insights for advancing explainable AI-generated video detection.</p> + https://arxiv.org/abs/2512.15693 + Wed, 17 Dec 2025 18:48:26 +0000 - MoRel: Long-Range Flicker-Free 4D Motion Modeling via Anchor Relay-based Bidirectional Blending with Hierarchical Densification - https://arxiv.org/abs/2512.09270 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.09270.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Sangwoon Kwak, Weeyoung Kwon, Jun Young Jeong, Geonho Kim, Won-Sik Cheong, Jihyong Oh</p><p><b>Upvotes:</b> 4</p><p><b>Summary:</b> Recent advances in 4D Gaussian Splatting (4DGS) have extended the high-speed rendering capability of 3D Gaussian Splatting (3DGS) into the temporal domain, enabling real-time rendering of dynamic scenes. However, one of the major remaining challenges lies in modeling long-range motion-contained dynamic videos, where a naive extension of existing methods leads to severe memory explosion, temporal flickering, and failure to handle appearing or disappearing occlusions over time. To address these challenges, we propose a novel 4DGS framework characterized by an Anchor Relay-based Bidirectional Blending (ARBB) mechanism, named MoRel, which enables temporally consistent and memory-efficient modeling of long-range dynamic scenes. Our method progressively constructs locally canonical anchor spaces at key-frame time index and models inter-frame deformations at the anchor level, enhancing temporal coherence. By learning bidirectional deformations between KfA and adaptively blending them through learnable opacity control, our approach mitigates temporal discontinuities and flickering artifacts. We further introduce a Feature-variance-guided Hierarchical Densification (FHD) scheme that effectively densifies KfA's while keeping rendering quality, based on an assigned level of feature-variance. To effectively evaluate our model's capability to handle real-world long-range 4D motion, we newly compose long-range 4D motion-contained dataset, called SelfCap_{LR}. It has larger average dynamic motion magnitude, captured at spatially wider spaces, compared to previous dynamic video datasets. Overall, our MoRel achieves temporally coherent and flicker-free long-range 4D reconstruction while maintaining bounded memory usage, demonstrating both scalability and efficiency in dynamic Gaussian-based representations.</p> - https://arxiv.org/abs/2512.09270 - Wed, 10 Dec 2025 02:49:09 +0000 + End-to-End Training for Autoregressive Video Diffusion via Self-Resampling + https://arxiv.org/abs/2512.15702 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.15702.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yuwei Guo, Ceyuan Yang, Hao He, Yang Zhao, Meng Wei, Zhenheng Yang, Weilin Huang, Dahua Lin</p><p><b>Upvotes:</b> 13</p><p><b>Summary:</b> Autoregressive video diffusion models hold promise for world simulation but are vulnerable to exposure bias arising from the train-test mismatch. While recent works address this via post-training, they typically rely on a bidirectional teacher model or online discriminator. To achieve an end-to-end solution, we introduce Resampling Forcing, a teacher-free framework that enables training autoregressive video models from scratch and at scale. Central to our approach is a self-resampling scheme that simulates inference-time model errors on history frames during training. Conditioned on these degraded histories, a sparse causal mask enforces temporal causality while enabling parallel training with frame-level diffusion loss. To facilitate efficient long-horizon generation, we further introduce history routing, a parameter-free mechanism that dynamically retrieves the top-k most relevant history frames for each query. Experiments demonstrate that our approach achieves performance comparable to distillation-based baselines while exhibiting superior temporal consistency on longer videos owing to native-length training.</p> + https://arxiv.org/abs/2512.15702 + Wed, 17 Dec 2025 18:53:29 +0000 - StereoWorld: Geometry-Aware Monocular-to-Stereo Video Generation - https://arxiv.org/abs/2512.09363 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.09363.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Ke Xing, Longfei Li, Yuyang Yin, Hanwen Liang, Guixun Luo, Chen Fang, Jue Wang, Konstantinos N. Plataniotis, Xiaojie Jin, Yao Zhao, Yunchao Wei</p><p><b>Upvotes:</b> 66</p><p><b>Summary:</b> The growing adoption of XR devices has fueled strong demand for high-quality stereo video, yet its production remains costly and artifact-prone. To address this challenge, we present StereoWorld, an end-to-end framework that repurposes a pretrained video generator for high-fidelity monocular-to-stereo video generation. Our framework jointly conditions the model on the monocular video input while explicitly supervising the generation with a geometry-aware regularization to ensure 3D structural fidelity. A spatio-temporal tiling scheme is further integrated to enable efficient, high-resolution synthesis. To enable large-scale training and evaluation, we curate a high-definition stereo video dataset containing over 11M frames aligned to natural human interpupillary distance (IPD). Extensive experiments demonstrate that StereoWorld substantially outperforms prior methods, generating stereo videos with superior visual fidelity and geometric consistency. The project webpage is available at https://ke-xing.github.io/StereoWorld/.</p> - https://arxiv.org/abs/2512.09363 - Wed, 10 Dec 2025 06:50:16 +0000 + DiffusionVL: Translating Any Autoregressive Models into Diffusion Vision Language Models + https://arxiv.org/abs/2512.15713 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.15713.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Lunbin Zeng, Jingfeng Yao, Bencheng Liao, Hongyuan Tao, Wenyu Liu, Xinggang Wang</p><p><b>Upvotes:</b> 13</p><p><b>Summary:</b> In recent multimodal research, the diffusion paradigm has emerged as a promising alternative to the autoregressive paradigm (AR), owing to its unique decoding advantages. However, due to the capability limitations of the base diffusion language model, the performance of the diffusion vision language model (dVLM) still lags significantly behind that of mainstream models. This leads to a simple yet fundamental question: Is it possible to construct dVLMs based on existing powerful AR models? In response, we propose DiffusionVL, a dVLM family that could be translated from any powerful AR models. Through simple fine-tuning, we successfully adapt AR pre-trained models into the diffusion paradigm. This approach yields two key observations: (1) The paradigm shift from AR-based multimodal models to diffusion is remarkably effective. (2) Direct conversion of an AR language model to a dVLM is also feasible, achieving performance competitive with LLaVA-style visual-instruction-tuning. Further, we introduce a block-decoding design into dVLMs that supports arbitrary-length generation and KV cache reuse, achieving a significant inference speedup. We conduct a large number of experiments. Despite training with less than 5% of the data required by prior methods, DiffusionVL achieves a comprehensive performance improvement-a 34.4% gain on the MMMU-Pro (vision) bench and 37.5% gain on the MME (Cog.) bench-alongside a 2x inference speedup. The model and code are released at https://github.com/hustvl/DiffusionVL.</p> + https://arxiv.org/abs/2512.15713 + Wed, 17 Dec 2025 18:59:55 +0000 - H2R-Grounder: A Paired-Data-Free Paradigm for Translating Human Interaction Videos into Physically Grounded Robot Videos - https://arxiv.org/abs/2512.09406 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.09406.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Hai Ci, Xiaokang Liu, Pei Yang, Yiren Song, Mike Zheng Shou</p><p><b>Upvotes:</b> 3</p><p><b>Summary:</b> Robots that learn manipulation skills from everyday human videos could acquire broad capabilities without tedious robot data collection. We propose a video-to-video translation framework that converts ordinary human-object interaction videos into motion-consistent robot manipulation videos with realistic, physically grounded interactions. Our approach does not require any paired human-robot videos for training only a set of unpaired robot videos, making the system easy to scale. We introduce a transferable representation that bridges the embodiment gap: by inpainting the robot arm in training videos to obtain a clean background and overlaying a simple visual cue (a marker and arrow indicating the gripper's position and orientation), we can condition a generative model to insert the robot arm back into the scene. At test time, we apply the same process to human videos (inpainting the person and overlaying human pose cues) and generate high-quality robot videos that mimic the human's actions. We fine-tune a SOTA video diffusion model (Wan 2.2) in an in-context learning manner to ensure temporal coherence and leveraging of its rich prior knowledge. Empirical results demonstrate that our approach achieves significantly more realistic and grounded robot motions compared to baselines, pointing to a promising direction for scaling up robot learning from unlabeled human videos. Project page: https://showlab.github.io/H2R-Grounder/</p> - https://arxiv.org/abs/2512.09406 - Wed, 10 Dec 2025 07:59:45 +0000 + In Pursuit of Pixel Supervision for Visual Pre-training + https://arxiv.org/abs/2512.15715 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.15715.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Lihe Yang, Shang-Wen Li, Yang Li, Xinjie Lei, Dong Wang, Abdelrahman Mohamed, Hengshuang Zhao, Hu Xu</p><p><b>Upvotes:</b> 7</p><p><b>Summary:</b> At the most basic level, pixels are the source of the visual information through which we perceive the world. Pixels contain information at all levels, ranging from low-level attributes to high-level concepts. Autoencoders represent a classical and long-standing paradigm for learning representations from pixels or other raw inputs. In this work, we demonstrate that autoencoder-based self-supervised learning remains competitive today and can produce strong representations for downstream tasks, while remaining simple, stable, and efficient. Our model, codenamed "Pixio", is an enhanced masked autoencoder (MAE) with more challenging pre-training tasks and more capable architectures. The model is trained on 2B web-crawled images with a self-curation strategy with minimal human curation. Pixio performs competitively across a wide range of downstream tasks in the wild, including monocular depth estimation (e.g., Depth Anything), feed-forward 3D reconstruction (i.e., MapAnything), semantic segmentation, and robot learning, outperforming or matching DINOv3 trained at similar scales. Our results suggest that pixel-space self-supervised learning can serve as a promising alternative and a complement to latent-space approaches.</p> + https://arxiv.org/abs/2512.15715 + Wed, 17 Dec 2025 18:59:58 +0000 - Rethinking Chain-of-Thought Reasoning for Videos - https://arxiv.org/abs/2512.09616 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.09616.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yiwu Zhong, Zi-Yuan Hu, Yin Li, Liwei Wang</p><p><b>Upvotes:</b> 15</p><p><b>Summary:</b> Chain-of-thought (CoT) reasoning has been highly successful in solving complex tasks in natural language processing, and recent multimodal large language models (MLLMs) have extended this paradigm to video reasoning. However, these models typically build on lengthy reasoning chains and large numbers of input visual tokens. Motivated by empirical observations from our benchmark study, we hypothesize that concise reasoning combined with a reduced set of visual tokens can be sufficient for effective video reasoning. To evaluate this hypothesis, we design and validate an efficient post-training and inference framework that enhances a video MLLM's reasoning capability. Our framework enables models to operate on compressed visual tokens and generate brief reasoning traces prior to answering. The resulting models achieve substantially improved inference efficiency, deliver competitive performance across diverse benchmarks, and avoid reliance on manual CoT annotations or supervised fine-tuning. Collectively, our results suggest that long, human-like CoT reasoning may not be necessary for general video reasoning, and that concise reasoning can be both effective and efficient. Our code will be released at https://github.com/LaVi-Lab/Rethink_CoT_Video.</p> - https://arxiv.org/abs/2512.09616 - Wed, 10 Dec 2025 13:05:55 +0000 + Adaptation of Agentic AI + https://arxiv.org/abs/2512.16301 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.16301.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Pengcheng Jiang, Jiacheng Lin, Zhiyi Shi, Zifeng Wang, Luxi He, Yichen Wu, Ming Zhong, Peiyang Song, Qizheng Zhang, Heng Wang, Xueqiang Xu, Hanwen Xu, Pengrui Han, Dylan Zhang, Jiashuo Sun, Chaoqi Yang, Kun Qian, Tian Wang, Changran Hu, Manling Li, Quanzheng Li, Hao Peng, Sheng Wang, Jingbo Shang, Chao Zhang, Jiaxuan You, Liyuan Liu, Pan Lu, Yu Zhang, Heng Ji, Yejin Choi, Dawn Song, Jimeng Sun, Jiawei Han</p><p><b>Upvotes:</b> 80</p><p><b>Summary:</b> Cutting-edge agentic AI systems are built on foundation models that can be adapted to plan, reason, and interact with external tools to perform increasingly complex and specialized tasks. As these systems grow in capability and scope, adaptation becomes a central mechanism for improving performance, reliability, and generalization. In this paper, we unify the rapidly expanding research landscape into a systematic framework that spans both agent adaptations and tool adaptations. We further decompose these into tool-execution-signaled and agent-output-signaled forms of agent adaptation, as well as agent-agnostic and agent-supervised forms of tool adaptation. We demonstrate that this framework helps clarify the design space of adaptation strategies in agentic AI, makes their trade-offs explicit, and provides practical guidance for selecting or switching among strategies during system design. We then review the representative approaches in each category, analyze their strengths and limitations, and highlight key open challenges and future opportunities. Overall, this paper aims to offer a conceptual foundation and practical roadmap for researchers and practitioners seeking to build more capable, efficient, and reliable agentic AI systems.</p> + https://arxiv.org/abs/2512.16301 + Thu, 18 Dec 2025 08:38:51 +0000 - Composing Concepts from Images and Videos via Concept-prompt Binding - https://arxiv.org/abs/2512.09824 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.09824.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Xianghao Kong, Zeyu Zhang, Yuwei Guo, Zhuoran Zhao, Songchun Zhang, Anyi Rao</p><p><b>Upvotes:</b> 26</p><p><b>Summary:</b> Visual concept composition, which aims to integrate different elements from images and videos into a single, coherent visual output, still falls short in accurately extracting complex concepts from visual inputs and flexibly combining concepts from both images and videos. We introduce Bind & Compose, a one-shot method that enables flexible visual concept composition by binding visual concepts with corresponding prompt tokens and composing the target prompt with bound tokens from various sources. It adopts a hierarchical binder structure for cross-attention conditioning in Diffusion Transformers to encode visual concepts into corresponding prompt tokens for accurate decomposition of complex visual concepts. To improve concept-token binding accuracy, we design a Diversify-and-Absorb Mechanism that uses an extra absorbent token to eliminate the impact of concept-irrelevant details when training with diversified prompts. To enhance the compatibility between image and video concepts, we present a Temporal Disentanglement Strategy that decouples the training process of video concepts into two stages with a dual-branch binder structure for temporal modeling. Evaluations demonstrate that our method achieves superior concept consistency, prompt fidelity, and motion quality over existing approaches, opening up new possibilities for visual creativity.</p> - https://arxiv.org/abs/2512.09824 - Wed, 10 Dec 2025 16:57:31 +0000 + VenusBench-GD: A Comprehensive Multi-Platform GUI Benchmark for Diverse Grounding Tasks + https://arxiv.org/abs/2512.16501 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.16501.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Beitong Zhou, Zhexiao Huang, Yuan Guo, Zhangxuan Gu, Tianyu Xia, Zichen Luo, Fei Tang, Dehan Kong, Yanyi Shang, Suling Ou, Zhenlin Guo, Changhua Meng, Shuheng Shen</p><p><b>Upvotes:</b> 8</p><p><b>Summary:</b> GUI grounding is a critical component in building capable GUI agents. However, existing grounding benchmarks suffer from significant limitations: they either provide insufficient data volume and narrow domain coverage, or focus excessively on a single platform and require highly specialized domain knowledge. In this work, we present VenusBench-GD, a comprehensive, bilingual benchmark for GUI grounding that spans multiple platforms, enabling hierarchical evaluation for real-word applications. VenusBench-GD contributes as follows: (i) we introduce a large-scale, cross-platform benchmark with extensive coverage of applications, diverse UI elements, and rich annotated data, (ii) we establish a high-quality data construction pipeline for grounding tasks, achieving higher annotation accuracy than existing benchmarks, and (iii) we extend the scope of element grounding by proposing a hierarchical task taxonomy that divides grounding into basic and advanced categories, encompassing six distinct subtasks designed to evaluate models from complementary perspectives. Our experimental findings reveal critical insights: general-purpose multimodal models now match or even surpass specialized GUI models on basic grounding tasks. In contrast, advanced tasks, still favor GUI-specialized models, though they exhibit significant overfitting and poor robustness. These results underscore the necessity of comprehensive, multi-tiered evaluation frameworks.</p> + https://arxiv.org/abs/2512.16501 + Thu, 18 Dec 2025 13:09:09 +0000 - UniUGP: Unifying Understanding, Generation, and Planing For End-to-end Autonomous Driving - https://arxiv.org/abs/2512.09864 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.09864.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Hao Lu, Ziyang Liu, Guangfeng Jiang, Yuanfei Luo, Sheng Chen, Yangang Zhang, Ying-Cong Chen</p><p><b>Upvotes:</b> 10</p><p><b>Summary:</b> Autonomous driving (AD) systems struggle in long-tail scenarios due to limited world knowledge and weak visual dynamic modeling. Existing vision-language-action (VLA)-based methods cannot leverage unlabeled videos for visual causal learning, while world model-based methods lack reasoning capabilities from large language models. In this paper, we construct multiple specialized datasets providing reasoning and planning annotations for complex scenarios. Then, a unified Understanding-Generation-Planning framework, named UniUGP, is proposed to synergize scene reasoning, future video generation, and trajectory planning through a hybrid expert architecture. By integrating pre-trained VLMs and video generation models, UniUGP leverages visual dynamics and semantic reasoning to enhance planning performance. Taking multi-frame observations and language instructions as input, it produces interpretable chain-of-thought reasoning, physically consistent trajectories, and coherent future videos. We introduce a four-stage training strategy that progressively builds these capabilities across multiple existing AD datasets, along with the proposed specialized datasets. Experiments demonstrate state-of-the-art performance in perception, reasoning, and decision-making, with superior generalization to challenging long-tail situations.</p> - https://arxiv.org/abs/2512.09864 - Wed, 10 Dec 2025 17:50:29 +0000 + N3D-VLM: Native 3D Grounding Enables Accurate Spatial Reasoning in Vision-Language Models + https://arxiv.org/abs/2512.16561 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.16561.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yuxin Wang, Lei Ke, Boqiang Zhang, Tianyuan Qu, Hanxun Yu, Zhenpeng Huang, Meng Yu, Dan Xu, Dong Yu</p><p><b>Upvotes:</b> 18</p><p><b>Summary:</b> While current multimodal models can answer questions based on 2D images, they lack intrinsic 3D object perception, limiting their ability to comprehend spatial relationships and depth cues in 3D scenes. In this work, we propose N3D-VLM, a novel unified framework that seamlessly integrates native 3D object perception with 3D-aware visual reasoning, enabling both precise 3D grounding and interpretable spatial understanding. Unlike conventional end-to-end models that directly predict answers from RGB/RGB-D inputs, our approach equips the model with native 3D object perception capabilities, enabling it to directly localize objects in 3D space based on textual descriptions. Building upon accurate 3D object localization, the model further performs explicit reasoning in 3D, achieving more interpretable and structured spatial understanding. To support robust training for these capabilities, we develop a scalable data construction pipeline that leverages depth estimation to lift large-scale 2D annotations into 3D space, significantly increasing the diversity and coverage for 3D object grounding data, yielding over six times larger than the largest existing single-image 3D detection dataset. Moreover, the pipeline generates spatial question-answering datasets that target chain-of-thought (CoT) reasoning in 3D, facilitating joint training for both 3D object localization and 3D spatial reasoning. Experimental results demonstrate that our unified framework not only achieves state-of-the-art performance on 3D grounding tasks, but also consistently surpasses existing methods in 3D spatial reasoning in vision-language model.</p> + https://arxiv.org/abs/2512.16561 + Thu, 18 Dec 2025 14:03:44 +0000 - ReViSE: Towards Reason-Informed Video Editing in Unified Models with Self-Reflective Learning - https://arxiv.org/abs/2512.09924 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.09924.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Xinyu Liu, Hangjie Yuan, Yujie Wei, Jiazheng Xing, Yujin Han, Jiahao Pan, Yanbiao Ma, Chi-Min Chan, Kang Zhao, Shiwei Zhang, Wenhan Luo, Yike Guo</p><p><b>Upvotes:</b> 3</p><p><b>Summary:</b> Video unified models exhibit strong capabilities in understanding and generation, yet they struggle with reason-informed visual editing even when equipped with powerful internal vision-language models (VLMs). We attribute this gap to two factors: 1) existing datasets are inadequate for training and evaluating reasoning-aware video editing, and 2) an inherent disconnect between the models' reasoning and editing capabilities, which prevents the rich understanding from effectively instructing the editing process. Bridging this gap requires an integrated framework that connects reasoning with visual transformation. To address this gap, we introduce the Reason-Informed Video Editing (RVE) task, which requires reasoning about physical plausibility and causal dynamics during editing. To support systematic evaluation, we construct RVE-Bench, a comprehensive benchmark with two complementary subsets: Reasoning-Informed Video Editing and In-Context Video Generation. These subsets cover diverse reasoning dimensions and real-world editing scenarios. Building upon this foundation, we propose the ReViSE, a Self-Reflective Reasoning (SRF) framework that unifies generation and evaluation within a single architecture. The model's internal VLM provides intrinsic feedback by assessing whether the edited video logically satisfies the given instruction. The differential feedback that refines the generator's reasoning behavior during training. Extensive experiments on RVE-Bench demonstrate that ReViSE significantly enhances editing accuracy and visual fidelity, achieving a 32% improvement of the Overall score in the reasoning-informed video editing subset over state-of-the-art methods.</p> - https://arxiv.org/abs/2512.09924 - Wed, 10 Dec 2025 18:57:09 +0000 + DeContext as Defense: Safe Image Editing in Diffusion Transformers + https://arxiv.org/abs/2512.16625 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.16625.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Linghui Shen, Mingyue Cui, Xingyi Yang</p><p><b>Upvotes:</b> 24</p><p><b>Summary:</b> In-context diffusion models allow users to modify images with remarkable ease and realism. However, the same power raises serious privacy concerns: personal images can be easily manipulated for identity impersonation, misinformation, or other malicious uses, all without the owner's consent. While prior work has explored input perturbations to protect against misuse in personalized text-to-image generation, the robustness of modern, large-scale in-context DiT-based models remains largely unexamined. In this paper, we propose DeContext, a new method to safeguard input images from unauthorized in-context editing. Our key insight is that contextual information from the source image propagates to the output primarily through multimodal attention layers. By injecting small, targeted perturbations that weaken these cross-attention pathways, DeContext breaks this flow, effectively decouples the link between input and output. This simple defense is both efficient and robust. We further show that early denoising steps and specific transformer blocks dominate context propagation, which allows us to concentrate perturbations where they matter most. Experiments on Flux Kontext and Step1X-Edit show that DeContext consistently blocks unwanted image edits while preserving visual quality. These results highlight the effectiveness of attention-based perturbations as a powerful defense against image manipulation.</p> + https://arxiv.org/abs/2512.16625 + Thu, 18 Dec 2025 15:01:44 +0000 - HiF-VLA: Hindsight, Insight and Foresight through Motion Representation for Vision-Language-Action Models - https://arxiv.org/abs/2512.09928 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.09928.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Minghui Lin, Pengxiang Ding, Shu Wang, Zifeng Zhuang, Yang Liu, Xinyang Tong, Wenxuan Song, Shangke Lyu, Siteng Huang, Donglin Wang</p><p><b>Upvotes:</b> 10</p><p><b>Summary:</b> Vision-Language-Action (VLA) models have recently enabled robotic manipulation by grounding visual and linguistic cues into actions. However, most VLAs assume the Markov property, relying only on the current observation and thus suffering from temporal myopia that degrades long-horizon coherence. In this work, we view motion as a more compact and informative representation of temporal context and world dynamics, capturing inter-state changes while filtering static pixel-level noise. Building on this idea, we propose HiF-VLA (Hindsight, Insight, and Foresight for VLAs), a unified framework that leverages motion for bidirectional temporal reasoning. HiF-VLA encodes past dynamics through hindsight priors, anticipates future motion via foresight reasoning, and integrates both through a hindsight-modulated joint expert to enable a ''think-while-acting'' paradigm for long-horizon manipulation. As a result, HiF-VLA surpasses strong baselines on LIBERO-Long and CALVIN ABC-D benchmarks, while incurring negligible additional inference latency. Furthermore, HiF-VLA achieves substantial improvements in real-world long-horizon manipulation tasks, demonstrating its broad effectiveness in practical robotic settings.</p> - https://arxiv.org/abs/2512.09928 - Wed, 10 Dec 2025 18:59:32 +0000 + REGLUE Your Latents with Global and Local Semantics for Entangled Diffusion + https://arxiv.org/abs/2512.16636 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.16636.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Giorgos Petsangourakis, Christos Sgouropoulos, Bill Psomas, Theodoros Giannakopoulos, Giorgos Sfikas, Ioannis Kakogeorgiou</p><p><b>Upvotes:</b> 23</p><p><b>Summary:</b> Latent diffusion models (LDMs) achieve state-of-the-art image synthesis, yet their reconstruction-style denoising objective provides only indirect semantic supervision: high-level semantics emerge slowly, requiring longer training and limiting sample quality. Recent works inject semantics from Vision Foundation Models (VFMs) either externally via representation alignment or internally by jointly modeling only a narrow slice of VFM features inside the diffusion process, under-utilizing the rich, nonlinear, multi-layer spatial semantics available. We introduce REGLUE (Representation Entanglement with Global-Local Unified Encoding), a unified latent diffusion framework that jointly models (i) VAE image latents, (ii) compact local (patch-level) VFM semantics, and (iii) a global (image-level) [CLS] token within a single SiT backbone. A lightweight convolutional semantic compressor nonlinearly aggregates multi-layer VFM features into a low-dimensional, spatially structured representation, which is entangled with the VAE latents in the diffusion process. An external alignment loss further regularizes internal representations toward frozen VFM targets. On ImageNet 256x256, REGLUE consistently improves FID and accelerates convergence over SiT-B/2 and SiT-XL/2 baselines, as well as over REPA, ReDi, and REG. Extensive experiments show that (a) spatial VFM semantics are crucial, (b) non-linear compression is key to unlocking their full benefit, and (c) global tokens and external alignment act as complementary, lightweight enhancements within our global-local-latent joint modeling framework. The code is available at https://github.com/giorgospets/reglue .</p> + https://arxiv.org/abs/2512.16636 + Thu, 18 Dec 2025 15:10:42 +0000 - MotionEdit: Benchmarking and Learning Motion-Centric Image Editing - https://arxiv.org/abs/2512.10284 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.10284.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yixin Wan, Lei Ke, Wenhao Yu, Kai-Wei Chang, Dong Yu</p><p><b>Upvotes:</b> 23</p><p><b>Summary:</b> We introduce MotionEdit, a novel dataset for motion-centric image editing-the task of modifying subject actions and interactions while preserving identity, structure, and physical plausibility. Unlike existing image editing datasets that focus on static appearance changes or contain only sparse, low-quality motion edits, MotionEdit provides high-fidelity image pairs depicting realistic motion transformations extracted and verified from continuous videos. This new task is not only scientifically challenging but also practically significant, powering downstream applications such as frame-controlled video synthesis and animation. To evaluate model performance on the novel task, we introduce MotionEdit-Bench, a benchmark that challenges models on motion-centric edits and measures model performance with generative, discriminative, and preference-based metrics. Benchmark results reveal that motion editing remains highly challenging for existing state-of-the-art diffusion-based editing models. To address this gap, we propose MotionNFT (Motion-guided Negative-aware Fine Tuning), a post-training framework that computes motion alignment rewards based on how well the motion flow between input and model-edited images matches the ground-truth motion, guiding models toward accurate motion transformations. Extensive experiments on FLUX.1 Kontext and Qwen-Image-Edit show that MotionNFT consistently improves editing quality and motion fidelity of both base models on the motion editing task without sacrificing general editing ability, demonstrating its effectiveness.</p> - https://arxiv.org/abs/2512.10284 - Thu, 11 Dec 2025 04:53:58 +0000 + JustRL: Scaling a 1.5B LLM with a Simple RL Recipe + https://arxiv.org/abs/2512.16649 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.16649.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Bingxiang He, Zekai Qu, Zeyuan Liu, Yinghao Chen, Yuxin Zuo, Cheng Qian, Kaiyan Zhang, Weize Chen, Chaojun Xiao, Ganqu Cui, Ning Ding, Zhiyuan Liu</p><p><b>Upvotes:</b> 18</p><p><b>Summary:</b> Recent advances in reinforcement learning for large language models have converged on increasing complexity: multi-stage training pipelines, dynamic hyperparameter schedules, and curriculum learning strategies. This raises a fundamental question: Is this complexity necessary? We present JustRL, a minimal approach using single-stage training with fixed hyperparameters that achieves state-of-the-art performance on two 1.5B reasoning models (54.9\% and 64.3\% average accuracy across nine mathematical benchmarks) while using 2times less compute than sophisticated approaches. The same hyperparameters transfer across both models without tuning, and training exhibits smooth, monotonic improvement over 4,000+ steps without the collapses or plateaus that typically motivate interventions. Critically, ablations reveal that adding ``standard tricks'' like explicit length penalties and robust verifiers may degrade performance by collapsing exploration. These results suggest that the field may be adding complexity to solve problems that disappear with a stable, scaled-up baseline. We release our models and code to establish a simple, validated baseline for the community.</p> + https://arxiv.org/abs/2512.16649 + Thu, 18 Dec 2025 15:21:25 +0000 - Tool-Augmented Spatiotemporal Reasoning for Streamlining Video Question Answering Task - https://arxiv.org/abs/2512.10359 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.10359.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Sunqi Fan, Jiashuo Cui, Meng-Hao Guo, Shuojin Yang</p><p><b>Upvotes:</b> 3</p><p><b>Summary:</b> Video Question Answering (VideoQA) task serves as a critical playground for evaluating whether foundation models can effectively perceive, understand, and reason about dynamic real-world scenarios. However, existing Multimodal Large Language Models (MLLMs) struggle with simultaneously modeling spatial relationships within video frames and understanding the causal dynamics of temporal evolution on complex and reasoning-intensive VideoQA task. In this work, we equip MLLM with a comprehensive and extensible Video Toolkit, to enhance MLLM's spatiotemporal reasoning capabilities and ensure the harmony between the quantity and diversity of tools. To better control the tool invocation sequence and avoid toolchain shortcut issues, we propose a Spatiotemporal Reasoning Framework (STAR) that strategically schedules temporal and spatial tools, thereby progressively localizing the key area in the video. Our STAR framework enhances GPT-4o using lightweight tools, achieving an 8.2% gain on VideoMME and 4.6% on LongVideoBench. We believe that our proposed Video Toolkit and STAR framework make an important step towards building autonomous and intelligent video analysis assistants. The code is publicly available at https://github.com/fansunqi/VideoTool.</p> - https://arxiv.org/abs/2512.10359 - Thu, 11 Dec 2025 07:17:57 +0000 + Vision-Language-Action Models for Autonomous Driving: Past, Present, and Future + https://arxiv.org/abs/2512.16760 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.16760.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Tianshuai Hu, Xiaolu Liu, Song Wang, Yiyao Zhu, Ao Liang, Lingdong Kong, Guoyang Zhao, Zeying Gong, Jun Cen, Zhiyu Huang, Xiaoshuai Hao, Linfeng Li, Hang Song, Xiangtai Li, Jun Ma, Shaojie Shen, Jianke Zhu, Dacheng Tao, Ziwei Liu, Junwei Liang</p><p><b>Upvotes:</b> 12</p><p><b>Summary:</b> Autonomous driving has long relied on modular "Perception-Decision-Action" pipelines, where hand-crafted interfaces and rule-based components often break down in complex or long-tailed scenarios. Their cascaded design further propagates perception errors, degrading downstream planning and control. Vision-Action (VA) models address some limitations by learning direct mappings from visual inputs to actions, but they remain opaque, sensitive to distribution shifts, and lack structured reasoning or instruction-following capabilities. Recent progress in Large Language Models (LLMs) and multimodal learning has motivated the emergence of Vision-Language-Action (VLA) frameworks, which integrate perception with language-grounded decision making. By unifying visual understanding, linguistic reasoning, and actionable outputs, VLAs offer a pathway toward more interpretable, generalizable, and human-aligned driving policies. This work provides a structured characterization of the emerging VLA landscape for autonomous driving. We trace the evolution from early VA approaches to modern VLA frameworks and organize existing methods into two principal paradigms: End-to-End VLA, which integrates perception, reasoning, and planning within a single model, and Dual-System VLA, which separates slow deliberation (via VLMs) from fast, safety-critical execution (via planners). Within these paradigms, we further distinguish subclasses such as textual vs. numerical action generators and explicit vs. implicit guidance mechanisms. We also summarize representative datasets and benchmarks for evaluating VLA-based driving systems and highlight key challenges and open directions, including robustness, interpretability, and instruction fidelity. Overall, this work aims to establish a coherent foundation for advancing human-compatible autonomous driving systems.</p> + https://arxiv.org/abs/2512.16760 + Thu, 18 Dec 2025 16:57:44 +0000 - Confucius Code Agent: An Open-sourced AI Software Engineer at Industrial Scale - https://arxiv.org/abs/2512.10398 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.10398.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zhaodong Wang, Zhenting Qi, Sherman Wong, Nathan Hu, Samuel Lin, Jun Ge, Erwin Gao, Yining Yang, Ben Maurer, Wenlin Chen, David Recordon, Yilun Du, Minlan Yu, Ying Zhang</p><p><b>Upvotes:</b> 2</p><p><b>Summary:</b> Real-world AI software engineering demands coding agents that can reason over massive repositories, maintain durable memory across and within long sessions, and robustly coordinate complex toolchains at test time. Existing open-source coding agents provide transparency but frequently fall short when pushed to these industrial-scale workloads, while proprietary coding agents offer strong practical performance but limited extensibility, interpretability, and controllability. We present the Confucius Code Agent (CCA), an open-sourced AI software engineer that can operate at an industrial scale. CCA is built atop the Confucius SDK, an open-sourced agent development platform designed around three complementary perspectives: Agent Experience (AX), User Experience (UX), and Developer Experience (DX). The SDK introduces a unified orchestrator with hierarchical working memory for long-context reasoning, a persistent note-taking system for cross-session continual learning, and a modular extension module for robust tool use. Moreover, a meta-agent automates the synthesis, evaluation, and refinement of agent configurations through a build-test-improve loop, enabling rapid agent development on new tasks, environments, and tool stacks. Instantiated on Confucius SDK with these mechanisms, CCA delivers strong performance on real-world software engineering tasks. On SWE-Bench-Pro, CCA achieves a state-of-the-art Resolve@1 performance of 54.3%, substantially improving over prior coding agents. Together, the Confucius SDK and CCA provide a transparent, extensible, and reproducible foundation for AI agents, bridge gaps between research prototypes and production-grade systems, and support agent development and deployment at industrial scale.</p> - https://arxiv.org/abs/2512.10398 - Thu, 11 Dec 2025 08:05:58 +0000 + Kling-Omni Technical Report + https://arxiv.org/abs/2512.16776 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.16776.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Kling Team, Jialu Chen, Yuanzheng Ci, Xiangyu Du, Zipeng Feng, Kun Gai, Sainan Guo, Feng Han, Jingbin He, Kang He, Xiao Hu, Xiaohua Hu, Boyuan Jiang, Fangyuan Kong, Hang Li, Jie Li, Qingyu Li, Shen Li, Xiaohan Li, Yan Li, Jiajun Liang, Borui Liao, Yiqiao Liao, Weihong Lin, Quande Liu, Xiaokun Liu, Yilun Liu, Yuliang Liu, Shun Lu, Hangyu Mao, Yunyao Mao, Haodong Ouyang, Wenyu Qin, Wanqi Shi, Xiaoyu Shi, Lianghao Su, Haozhi Sun, Peiqin Sun, Pengfei Wan, Chao Wang, Chenyu Wang, Meng Wang, Qiulin Wang, Runqi Wang, Xintao Wang, Xuebo Wang, Zekun Wang, Min Wei, Tiancheng Wen, Guohao Wu, Xiaoshi Wu, Zhenhua Wu, Da Xie, Yingtong Xiong, Yulong Xu, Sile Yang, Zikang Yang, Weicai Ye, Ziyang Yuan, Shenglong Zhang, Shuaiyu Zhang, Yuanxing Zhang, Yufan Zhang, Wenzheng Zhao, Ruiliang Zhou, Yan Zhou, Guosheng Zhu, Yongjie Zhu</p><p><b>Upvotes:</b> 135</p><p><b>Summary:</b> We present Kling-Omni, a generalist generative framework designed to synthesize high-fidelity videos directly from multimodal visual language inputs. Adopting an end-to-end perspective, Kling-Omni bridges the functional separation among diverse video generation, editing, and intelligent reasoning tasks, integrating them into a holistic system. Unlike disjointed pipeline approaches, Kling-Omni supports a diverse range of user inputs, including text instructions, reference images, and video contexts, processing them into a unified multimodal representation to deliver cinematic-quality and highly-intelligent video content creation. To support these capabilities, we constructed a comprehensive data system that serves as the foundation for multimodal video creation. The framework is further empowered by efficient large-scale pre-training strategies and infrastructure optimizations for inference. Comprehensive evaluations reveal that Kling-Omni demonstrates exceptional capabilities in in-context generation, reasoning-based editing, and multimodal instruction following. Moving beyond a content creation tool, we believe Kling-Omni is a pivotal advancement toward multimodal world simulators capable of perceiving, reasoning, generating and interacting with the dynamic and complex worlds.</p> + https://arxiv.org/abs/2512.16776 + Thu, 18 Dec 2025 17:08:12 +0000 - T-pro 2.0: An Efficient Russian Hybrid-Reasoning Model and Playground - https://arxiv.org/abs/2512.10430 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.10430.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Dmitrii Stoianov, Danil Taranets, Olga Tsymboi, Ramil Latypov, Almaz Dautov, Vladislav Kruglikov, Nikita Surkov, German Abramov, Pavel Gein, Dmitry Abulkhanov, Mikhail Gashkov, Viktor Zelenkovskiy, Artem Batalov, Aleksandr Medvedev, Anatolii Potapov</p><p><b>Upvotes:</b> 74</p><p><b>Summary:</b> We introduce T-pro 2.0, an open-weight Russian LLM for hybrid reasoning and efficient inference. The model supports direct answering and reasoning-trace generation, using a Cyrillic-dense tokenizer and an adapted EAGLE speculative-decoding pipeline to reduce latency. To enable reproducible and extensible research, we release the model weights, the T-Wix 500k instruction corpus, the T-Math reasoning benchmark, and the EAGLE weights on Hugging Face. These resources allow users to study Russian-language reasoning and to extend or adapt both the model and the inference pipeline. A public web demo exposes reasoning and non-reasoning modes and illustrates the speedups achieved by our inference stack across domains. T-pro 2.0 thus serves as an accessible open system for building and evaluating efficient, practical Russian LLM applications.</p> - https://arxiv.org/abs/2512.10430 - Thu, 11 Dec 2025 08:40:10 +0000 + RePlan: Reasoning-guided Region Planning for Complex Instruction-based Image Editing + https://arxiv.org/abs/2512.16864 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.16864.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Tianyuan Qu, Lei Ke, Xiaohang Zhan, Longxiang Tang, Yuqi Liu, Bohao Peng, Bei Yu, Dong Yu, Jiaya Jia</p><p><b>Upvotes:</b> 10</p><p><b>Summary:</b> Instruction-based image editing enables natural-language control over visual modifications, yet existing models falter under Instruction-Visual Complexity (IV-Complexity), where intricate instructions meet cluttered or ambiguous scenes. We introduce RePlan (Region-aligned Planning), a plan-then-execute framework that couples a vision-language planner with a diffusion editor. The planner decomposes instructions via step-by-step reasoning and explicitly grounds them to target regions; the editor then applies changes using a training-free attention-region injection mechanism, enabling precise, parallel multi-region edits without iterative inpainting. To strengthen planning, we apply GRPO-based reinforcement learning using 1K instruction-only examples, yielding substantial gains in reasoning fidelity and format reliability. We further present IV-Edit, a benchmark focused on fine-grained grounding and knowledge-intensive edits. Across IV-Complex settings, RePlan consistently outperforms strong baselines trained on far larger datasets, improving regional precision and overall fidelity. Our project page: https://replan-iv-edit.github.io</p> + https://arxiv.org/abs/2512.16864 + Thu, 18 Dec 2025 18:34:23 +0000 - Achieving Olympia-Level Geometry Large Language Model Agent via Complexity Boosting Reinforcement Learning - https://arxiv.org/abs/2512.10534 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.10534.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Haiteng Zhao, Junhao Shen, Yiming Zhang, Songyang Gao, Kuikun Liu, Tianyou Ma, Fan Zheng, Dahua Lin, Wenwei Zhang, Kai Chen</p><p><b>Upvotes:</b> 27</p><p><b>Summary:</b> Large language model (LLM) agents exhibit strong mathematical problem-solving abilities and can even solve International Mathematical Olympiad (IMO) level problems with the assistance of formal proof systems. However, due to weak heuristics for auxiliary constructions, AI for geometry problem solving remains dominated by expert models such as AlphaGeometry 2, which rely heavily on large-scale data synthesis and search for both training and evaluation. In this work, we make the first attempt to build a medalist-level LLM agent for geometry and present InternGeometry. InternGeometry overcomes the heuristic limitations in geometry by iteratively proposing propositions and auxiliary constructions, verifying them with a symbolic engine, and reflecting on the engine's feedback to guide subsequent proposals. A dynamic memory mechanism enables InternGeometry to conduct more than two hundred interactions with the symbolic engine per problem. To further accelerate learning, we introduce Complexity-Boosting Reinforcement Learning (CBRL), which gradually increases the complexity of synthesized problems across training stages. Built on InternThinker-32B, InternGeometry solves 44 of 50 IMO geometry problems (2000-2024), exceeding the average gold medalist score (40.9), using only 13K training examples, just 0.004% of the data used by AlphaGeometry 2, demonstrating the potential of LLM agents on expert-level geometry tasks. InternGeometry can also propose novel auxiliary constructions for IMO problems that do not appear in human solutions. We will release the model, data, and symbolic engine to support future research.</p> - https://arxiv.org/abs/2512.10534 - Thu, 11 Dec 2025 11:05:04 +0000 + Multimodal RewardBench 2: Evaluating Omni Reward Models for Interleaved Text and Image + https://arxiv.org/abs/2512.16899 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.16899.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yushi Hu, Reyhane Askari-Hemmat, Melissa Hall, Emily Dinan, Luke Zettlemoyer, Marjan Ghazvininejad</p><p><b>Upvotes:</b> 10</p><p><b>Summary:</b> Reward models (RMs) are essential for training large language models (LLMs), but remain underexplored for omni models that handle interleaved image and text sequences. We introduce Multimodal RewardBench 2 (MMRB2), the first comprehensive benchmark for reward models on multimodal understanding and (interleaved) generation. MMRB2 spans four tasks: text-to-image, image editing, interleaved generation, and multimodal reasoning ("thinking-with-images"), providing 1,000 expert-annotated preference pairs per task from 23 models and agents across 21 source tasks. MMRB2 is designed with: (1) practical but challenging prompts; (2) responses from state-of-the-art models and agents; and (3) preference pairs with strong human-expert consensus, curated via an ensemble filtering strategy. Using MMRB2, we study existing judges for each subtask, including multimodal LLM-as-a-judge and models trained with human preferences. The latest Gemini 3 Pro attains 75-80% accuracy. GPT-5 and Gemini 2.5 Pro reach 66-75% accuracy, compared to >90% for humans, yet surpass the widely used GPT-4o (59%). The best performing open-source model Qwen3-VL-32B achieves similar accuracies as Gemini 2.5 Flash (64%). We also show that MMRB2 performance strongly correlates with downstream task success using Best-of-N sampling and conduct an in-depth analysis that shows key areas to improve the reward models going forward.</p> + https://arxiv.org/abs/2512.16899 + Thu, 18 Dec 2025 18:56:04 +0000 - Evaluating Gemini Robotics Policies in a Veo World Simulator - https://arxiv.org/abs/2512.10675 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.10675.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Gemini Robotics Team, Coline Devin, Yilun Du, Debidatta Dwibedi, Ruiqi Gao, Abhishek Jindal, Thomas Kipf, Sean Kirmani, Fangchen Liu, Anirudha Majumdar, Andrew Marmon, Carolina Parada, Yulia Rubanova, Dhruv Shah, Vikas Sindhwani, Jie Tan, Fei Xia, Ted Xiao, Sherry Yang, Wenhao Yu, Allan Zhou</p><p><b>Upvotes:</b> 10</p><p><b>Summary:</b> Generative world models hold significant potential for simulating interactions with visuomotor policies in varied environments. Frontier video models can enable generation of realistic observations and environment interactions in a scalable and general manner. However, the use of video models in robotics has been limited primarily to in-distribution evaluations, i.e., scenarios that are similar to ones used to train the policy or fine-tune the base video model. In this report, we demonstrate that video models can be used for the entire spectrum of policy evaluation use cases in robotics: from assessing nominal performance to out-of-distribution (OOD) generalization, and probing physical and semantic safety. We introduce a generative evaluation system built upon a frontier video foundation model (Veo). The system is optimized to support robot action conditioning and multi-view consistency, while integrating generative image-editing and multi-view completion to synthesize realistic variations of real-world scenes along multiple axes of generalization. We demonstrate that the system preserves the base capabilities of the video model to enable accurate simulation of scenes that have been edited to include novel interaction objects, novel visual backgrounds, and novel distractor objects. This fidelity enables accurately predicting the relative performance of different policies in both nominal and OOD conditions, determining the relative impact of different axes of generalization on policy performance, and performing red teaming of policies to expose behaviors that violate physical or semantic safety constraints. We validate these capabilities through 1600+ real-world evaluations of eight Gemini Robotics policy checkpoints and five tasks for a bimanual manipulator.</p> - https://arxiv.org/abs/2512.10675 - Thu, 11 Dec 2025 14:22:14 +0000 + FlashPortrait: 6x Faster Infinite Portrait Animation with Adaptive Latent Prediction + https://arxiv.org/abs/2512.16900 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.16900.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Shuyuan Tu, Yueming Pan, Yinming Huang, Xintong Han, Zhen Xing, Qi Dai, Kai Qiu, Chong Luo, Zuxuan Wu</p><p><b>Upvotes:</b> 9</p><p><b>Summary:</b> Current diffusion-based acceleration methods for long-portrait animation struggle to ensure identity (ID) consistency. This paper presents FlashPortrait, an end-to-end video diffusion transformer capable of synthesizing ID-preserving, infinite-length videos while achieving up to 6x acceleration in inference speed. In particular, FlashPortrait begins by computing the identity-agnostic facial expression features with an off-the-shelf extractor. It then introduces a Normalized Facial Expression Block to align facial features with diffusion latents by normalizing them with their respective means and variances, thereby improving identity stability in facial modeling. During inference, FlashPortrait adopts a dynamic sliding-window scheme with weighted blending in overlapping areas, ensuring smooth transitions and ID consistency in long animations. In each context window, based on the latent variation rate at particular timesteps and the derivative magnitude ratio among diffusion layers, FlashPortrait utilizes higher-order latent derivatives at the current timestep to directly predict latents at future timesteps, thereby skipping several denoising steps and achieving 6x speed acceleration. Experiments on benchmarks show the effectiveness of FlashPortrait both qualitatively and quantitatively.</p> + https://arxiv.org/abs/2512.16900 + Thu, 18 Dec 2025 18:56:05 +0000 - Long-horizon Reasoning Agent for Olympiad-Level Mathematical Problem Solving - https://arxiv.org/abs/2512.10739 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.10739.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Songyang Gao, Yuzhe Gu, Zijian Wu, Lingkai Kong, Wenwei Zhang, Zhongrui Cai, Fan Zheng, Tianyou Ma, Junhao Shen, Haiteng Zhao, Duanyang Zhang, Huilun Zhang, Kuikun Liu, Chengqi Lyu, Yanhui Duan, Chiyu Chen, Ningsheng Ma, Jianfei Gao, Han Lyu, Dahua Lin, Kai Chen</p><p><b>Upvotes:</b> 40</p><p><b>Summary:</b> Large language models (LLMs) have achieved significant progress in solving complex reasoning tasks by Reinforcement Learning with Verifiable Rewards (RLVR). This advancement is also inseparable from the oversight automated by reliable verifiers. However, current outcome-based verifiers (OVs) are unable to inspect the unreliable intermediate steps in the long reasoning chains of thought (CoTs). Meanwhile, current process-based verifiers (PVs) have difficulties in reliably detecting errors in the complex long CoTs, limited by the scarcity of high-quality annotations due to the prohibitive costs of human annotations. Therefore, we propose the Outcome-based Process Verifier (OPV), which verifies the rationale process of summarized outcomes from long CoTs to achieve both accurate and efficient verification and enable large-scale annotation. To empower the proposed verifier, we adopt an iterative active learning framework with expert annotations to progressively improve the verification capability of OPV with fewer annotation costs. Specifically, in each iteration, the most uncertain cases of the current best OPV are annotated and then subsequently used to train a new OPV through Rejection Fine-Tuning (RFT) and RLVR for the next round. Extensive experiments demonstrate OPV's superior performance and broad applicability. It achieves new state-of-the-art results on our held-out \thisbench, outperforming much larger open-source models such as Qwen3-Max-Preview with an F1 score of 83.1 compared to 76.3. Furthermore, OPV effectively detects false positives within synthetic dataset, closely align with expert assessment. When collaborating with policy models, OPV consistently yields performance gains, e.g., raising the accuracy of DeepSeek-R1-Distill-Qwen-32B from 55.2\% to 73.3\% on AIME2025 as the compute budget scales.</p> - https://arxiv.org/abs/2512.10739 - Thu, 11 Dec 2025 15:26:28 +0000 + Alchemist: Unlocking Efficiency in Text-to-Image Model Training via Meta-Gradient Data Selection + https://arxiv.org/abs/2512.16905 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.16905.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Kaixin Ding, Yang Zhou, Xi Chen, Miao Yang, Jiarong Ou, Rui Chen, Xin Tao, Hengshuang Zhao</p><p><b>Upvotes:</b> 24</p><p><b>Summary:</b> Recent advances in Text-to-Image (T2I) generative models, such as Imagen, Stable Diffusion, and FLUX, have led to remarkable improvements in visual quality. However, their performance is fundamentally limited by the quality of training data. Web-crawled and synthetic image datasets often contain low-quality or redundant samples, which lead to degraded visual fidelity, unstable training, and inefficient computation. Hence, effective data selection is crucial for improving data efficiency. Existing approaches rely on costly manual curation or heuristic scoring based on single-dimensional features in Text-to-Image data filtering. Although meta-learning based method has been explored in LLM, there is no adaptation for image modalities. To this end, we propose **Alchemist**, a meta-gradient-based framework to select a suitable subset from large-scale text-image data pairs. Our approach automatically learns to assess the influence of each sample by iteratively optimizing the model from a data-centric perspective. Alchemist consists of two key stages: data rating and data pruning. We train a lightweight rater to estimate each sample's influence based on gradient information, enhanced with multi-granularity perception. We then use the Shift-Gsampling strategy to select informative subsets for efficient model training. Alchemist is the first automatic, scalable, meta-gradient-based data selection framework for Text-to-Image model training. Experiments on both synthetic and web-crawled datasets demonstrate that Alchemist consistently improves visual quality and downstream performance. Training on an Alchemist-selected 50% of the data can outperform training on the full dataset.</p> + https://arxiv.org/abs/2512.16905 + Thu, 18 Dec 2025 18:57:58 +0000 - OPV: Outcome-based Process Verifier for Efficient Long Chain-of-Thought Verification - https://arxiv.org/abs/2512.10756 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.10756.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zijian Wu, Lingkai Kong, Wenwei Zhang, Songyang Gao, Yuzhe Gu, Zhongrui Cai, Tianyou Ma, Yuhong Liu, Zhi Wang, Runyuan Ma, Guangyu Wang, Wei Li, Conghui He, Dahua Lin, Kai Chen</p><p><b>Upvotes:</b> 31</p><p><b>Summary:</b> Large language models (LLMs) have achieved significant progress in solving complex reasoning tasks by Reinforcement Learning with Verifiable Rewards (RLVR). This advancement is also inseparable from the oversight automated by reliable verifiers. However, current outcome-based verifiers (OVs) are unable to inspect the unreliable intermediate steps in the long reasoning chains of thought (CoTs). Meanwhile, current process-based verifiers (PVs) have difficulties in reliably detecting errors in the complex long CoTs, limited by the scarcity of high-quality annotations due to the prohibitive costs of human annotations. Therefore, we propose the Outcome-based Process Verifier (OPV), which verifies the rationale process of summarized outcomes from long CoTs to achieve both accurate and efficient verification and enable large-scale annotation. To empower the proposed verifier, we adopt an iterative active learning framework with expert annotations to progressively improve the verification capability of OPV with fewer annotation costs. Specifically, in each iteration, the most uncertain cases of the current best OPV are annotated and then subsequently used to train a new OPV through Rejection Fine-Tuning (RFT) and RLVR for the next round. Extensive experiments demonstrate OPV's superior performance and broad applicability. It achieves new state-of-the-art results on our held-out OPV-Bench, outperforming much larger open-source models such as Qwen3-Max-Preview with an F1 score of 83.1 compared to 76.3. Furthermore, OPV effectively detects false positives within synthetic dataset, closely align with expert assessment. When collaborating with policy models, OPV consistently yields performance gains, e.g., raising the accuracy of DeepSeek-R1-Distill-Qwen-32B from 55.2% to 73.3% on AIME2025 as the compute budget scales.</p> - https://arxiv.org/abs/2512.10756 - Thu, 11 Dec 2025 15:47:38 +0000 + Exploration v.s. Exploitation: Rethinking RLVR through Clipping, Entropy, and Spurious Reward + https://arxiv.org/abs/2512.16912 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.16912.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Peter Chen, Xiaopeng Li, Ziniu Li, Wotao Yin, Xi Chen, Tianyi Lin</p><p><b>Upvotes:</b> 10</p><p><b>Summary:</b> This paper examines the exploration-exploitation trade-off in reinforcement learning with verifiable rewards (RLVR), a framework for improving the reasoning of Large Language Models (LLMs). Recent studies suggest that RLVR can elicit strong mathematical reasoning in LLMs through two seemingly paradoxical mechanisms: spurious rewards, which suppress exploitation by rewarding outcomes unrelated to the ground truth, and entropy minimization, which suppresses exploration by pushing the model toward more confident and deterministic outputs, highlighting a puzzling dynamic: both discouraging exploitation and discouraging exploration improve reasoning performance, yet the underlying principles that reconcile these effects remain poorly understood. We focus on two fundamental questions: (i) how policy entropy relates to performance, and (ii) whether spurious rewards yield gains, potentially through the interplay of clipping bias and model contamination. Our results show that clipping bias under spurious rewards reduces policy entropy, leading to more confident and deterministic outputs, while entropy minimization alone is insufficient for improvement. We further propose a reward-misalignment model explaining why spurious rewards can enhance performance beyond contaminated settings. Our findings clarify the mechanisms behind spurious-reward benefits and provide principles for more effective RLVR training.</p> + https://arxiv.org/abs/2512.16912 + Thu, 18 Dec 2025 18:59:27 +0000 - The FACTS Leaderboard: A Comprehensive Benchmark for Large Language Model Factuality - https://arxiv.org/abs/2512.10791 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.10791.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Aileen Cheng, Alon Jacovi, Amir Globerson, Ben Golan, Charles Kwong, Chris Alberti, Connie Tao, Eyal Ben-David, Gaurav Singh Tomar, Lukas Haas, Yonatan Bitton, Adam Bloniarz, Aijun Bai, Andrew Wang, Anfal Siddiqui, Arturo Bajuelos Castillo, Aviel Atias, Chang Liu, Corey Fry, Daniel Balle, Deepanway Ghosal, Doron Kukliansky, Dror Marcus, Elena Gribovskaya, Eran Ofek, Honglei Zhuang, Itay Laish, Jan Ackermann, Lily Wang, Meg Risdal, Megan Barnes, Michael Fink, Mohamed Amin, Moran Ambar, Natan Potikha, Nikita Gupta, Nitzan Katz, Noam Velan, Ofir Roval, Ori Ram, Polina Zablotskaia, Prathamesh Bang, Priyanka Agrawal, Rakesh Ghiya, Sanjay Ganapathy, Simon Baumgartner, Sofia Erell, Sushant Prakash, Thibault Sellam, Vikram Rao, Xuanhui Wang, Yaroslav Akulov, Yulong Yang, Zhen Yang, Zhixin Lai, Zhongru Wu, Anca Dragan, Avinatan Hassidim, Fernando Pereira, Slav Petrov, Srinivasan Venkatachary, Tulsee Doshi, Yossi Matias, Sasha Goldshtein, Dipanjan Das</p><p><b>Upvotes:</b> 4</p><p><b>Summary:</b> We introduce The FACTS Leaderboard, an online leaderboard suite and associated set of benchmarks that comprehensively evaluates the ability of language models to generate factually accurate text across diverse scenarios. The suite provides a holistic measure of factuality by aggregating the performance of models on four distinct sub-leaderboards: (1) FACTS Multimodal, which measures the factuality of responses to image-based questions; (2) FACTS Parametric, which assesses models' world knowledge by answering closed-book factoid questions from internal parameters; (3) FACTS Search, which evaluates factuality in information-seeking scenarios, where the model must use a search API; and (4) FACTS Grounding (v2), which evaluates whether long-form responses are grounded in provided documents, featuring significantly improved judge models. Each sub-leaderboard employs automated judge models to score model responses, and the final suite score is an average of the four components, designed to provide a robust and balanced assessment of a model's overall factuality. The FACTS Leaderboard Suite will be actively maintained, containing both public and private splits to allow for external participation while guarding its integrity. It can be found at https://www.kaggle.com/benchmarks/google/facts .</p> - https://arxiv.org/abs/2512.10791 - Thu, 11 Dec 2025 16:35:14 +0000 + Depth Any Panoramas: A Foundation Model for Panoramic Depth Estimation + https://arxiv.org/abs/2512.16913 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.16913.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Xin Lin, Meixi Song, Dizhe Zhang, Wenxuan Lu, Haodong Li, Bo Du, Ming-Hsuan Yang, Truong Nguyen, Lu Qi</p><p><b>Upvotes:</b> 30</p><p><b>Summary:</b> In this work, we present a panoramic metric depth foundation model that generalizes across diverse scene distances. We explore a data-in-the-loop paradigm from the view of both data construction and framework design. We collect a large-scale dataset by combining public datasets, high-quality synthetic data from our UE5 simulator and text-to-image models, and real panoramic images from the web. To reduce domain gaps between indoor/outdoor and synthetic/real data, we introduce a three-stage pseudo-label curation pipeline to generate reliable ground truth for unlabeled images. For the model, we adopt DINOv3-Large as the backbone for its strong pre-trained generalization, and introduce a plug-and-play range mask head, sharpness-centric optimization, and geometry-centric optimization to improve robustness to varying distances and enforce geometric consistency across views. Experiments on multiple benchmarks (e.g., Stanford2D3D, Matterport3D, and Deep360) demonstrate strong performance and zero-shot generalization, with particularly robust and stable metric predictions in diverse real-world scenes. The project page can be found at: https://insta360-research-team.github.io/DAP_website/ {https://insta360-research-team.github.io/DAP\_website/}</p> + https://arxiv.org/abs/2512.16913 + Thu, 18 Dec 2025 18:59:29 +0000 - From Macro to Micro: Benchmarking Microscopic Spatial Intelligence on Molecules via Vision-Language Models - https://arxiv.org/abs/2512.10867 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.10867.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zongzhao Li, Xiangzhe Kong, Jiahui Su, Zongyang Ma, Mingze Li, Songyou Li, Yuelin Zhang, Yu Rong, Tingyang Xu, Deli Zhao, Wenbing Huang</p><p><b>Upvotes:</b> 15</p><p><b>Summary:</b> This paper introduces the concept of Microscopic Spatial Intelligence (MiSI), the capability to perceive and reason about the spatial relationships of invisible microscopic entities, which is fundamental to scientific discovery. To assess the potential of Vision-Language Models (VLMs) in this domain, we propose a systematic benchmark framework MiSI-Bench. This framework features over 163,000 question-answer pairs and 587,000 images derived from approximately 4,000 molecular structures, covering nine complementary tasks that evaluate abilities ranging from elementary spatial transformations to complex relational identifications. Experimental results reveal that current state-of-the-art VLMs perform significantly below human level on this benchmark. However, a fine-tuned 7B model demonstrates substantial potential, even surpassing humans in spatial transformation tasks, while its poor performance in scientifically-grounded tasks like hydrogen bond recognition underscores the necessity of integrating explicit domain knowledge for progress toward scientific AGI. The datasets are available at https://huggingface.co/datasets/zongzhao/MiSI-bench.</p> - https://arxiv.org/abs/2512.10867 - Thu, 11 Dec 2025 18:00:21 +0000 + StereoPilot: Learning Unified and Efficient Stereo Conversion via Generative Priors + https://arxiv.org/abs/2512.16915 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.16915.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Guibao Shen, Yihua Du, Wenhang Ge, Jing He, Chirui Chang, Donghao Zhou, Zhen Yang, Luozhou Wang, Xin Tao, Ying-Cong Chen</p><p><b>Upvotes:</b> 36</p><p><b>Summary:</b> The rapid growth of stereoscopic displays, including VR headsets and 3D cinemas, has led to increasing demand for high-quality stereo video content. However, producing 3D videos remains costly and complex, while automatic Monocular-to-Stereo conversion is hindered by the limitations of the multi-stage ``Depth-Warp-Inpaint'' (DWI) pipeline. This paradigm suffers from error propagation, depth ambiguity, and format inconsistency between parallel and converged stereo configurations. To address these challenges, we introduce UniStereo, the first large-scale unified dataset for stereo video conversion, covering both stereo formats to enable fair benchmarking and robust model training. Building upon this dataset, we propose StereoPilot, an efficient feed-forward model that directly synthesizes the target view without relying on explicit depth maps or iterative diffusion sampling. Equipped with a learnable domain switcher and a cycle consistency loss, StereoPilot adapts seamlessly to different stereo formats and achieves improved consistency. Extensive experiments demonstrate that StereoPilot significantly outperforms state-of-the-art methods in both visual fidelity and computational efficiency. Project page: https://hit-perfect.github.io/StereoPilot/.</p> + https://arxiv.org/abs/2512.16915 + Thu, 18 Dec 2025 18:59:50 +0000 - MoCapAnything: Unified 3D Motion Capture for Arbitrary Skeletons from Monocular Videos - https://arxiv.org/abs/2512.10881 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.10881.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Kehong Gong, Zhengyu Wen, Weixia He, Mingxi Xu, Qi Wang, Ning Zhang, Zhengyu Li, Dongze Lian, Wei Zhao, Xiaoyu He, Mingyuan Zhang</p><p><b>Upvotes:</b> 25</p><p><b>Summary:</b> Motion capture now underpins content creation far beyond digital humans, yet most existing pipelines remain species- or template-specific. We formalize this gap as Category-Agnostic Motion Capture (CAMoCap): given a monocular video and an arbitrary rigged 3D asset as a prompt, the goal is to reconstruct a rotation-based animation such as BVH that directly drives the specific asset. We present MoCapAnything, a reference-guided, factorized framework that first predicts 3D joint trajectories and then recovers asset-specific rotations via constraint-aware inverse kinematics. The system contains three learnable modules and a lightweight IK stage: (1) a Reference Prompt Encoder that extracts per-joint queries from the asset's skeleton, mesh, and rendered images; (2) a Video Feature Extractor that computes dense visual descriptors and reconstructs a coarse 4D deforming mesh to bridge the gap between video and joint space; and (3) a Unified Motion Decoder that fuses these cues to produce temporally coherent trajectories. We also curate Truebones Zoo with 1038 motion clips, each providing a standardized skeleton-mesh-render triad. Experiments on both in-domain benchmarks and in-the-wild videos show that MoCapAnything delivers high-quality skeletal animations and exhibits meaningful cross-species retargeting across heterogeneous rigs, enabling scalable, prompt-driven 3D motion capture for arbitrary assets. Project page: https://animotionlab.github.io/MoCapAnything/</p> - https://arxiv.org/abs/2512.10881 - Thu, 11 Dec 2025 18:09:48 +0000 + AdaTooler-V: Adaptive Tool-Use for Images and Videos + https://arxiv.org/abs/2512.16918 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.16918.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Chaoyang Wang, Kaituo Feng, Dongyang Chen, Zhongyu Wang, Zhixun Li, Sicheng Gao, Meng Meng, Xu Zhou, Manyuan Zhang, Yuzhang Shang, Xiangyu Yue</p><p><b>Upvotes:</b> 11</p><p><b>Summary:</b> Recent advances have shown that multimodal large language models (MLLMs) benefit from multimodal interleaved chain-of-thought (CoT) with vision tool interactions. However, existing open-source models often exhibit blind tool-use reasoning patterns, invoking vision tools even when they are unnecessary, which significantly increases inference overhead and degrades model performance. To this end, we propose AdaTooler-V, an MLLM that performs adaptive tool-use by determining whether a visual problem truly requires tools. First, we introduce AT-GRPO, a reinforcement learning algorithm that adaptively adjusts reward scales based on the Tool Benefit Score of each sample, encouraging the model to invoke tools only when they provide genuine improvements. Moreover, we construct two datasets to support training: AdaTooler-V-CoT-100k for SFT cold start and AdaTooler-V-300k for RL with verifiable rewards across single-image, multi-image, and video data. Experiments across twelve benchmarks demonstrate the strong reasoning capability of AdaTooler-V, outperforming existing methods in diverse visual reasoning tasks. Notably, AdaTooler-V-7B achieves an accuracy of 89.8\% on the high-resolution benchmark V*, surpassing the commercial proprietary model GPT-4o and Gemini 1.5 Pro. All code, models, and data are released.</p> + https://arxiv.org/abs/2512.16918 + Thu, 18 Dec 2025 18:59:55 +0000 - DuetSVG: Unified Multimodal SVG Generation with Internal Visual Guidance - https://arxiv.org/abs/2512.10894 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.10894.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Peiying Zhang, Nanxuan Zhao, Matthew Fisher, Yiran Xu, Jing Liao, Difan Liu</p><p><b>Upvotes:</b> 2</p><p><b>Summary:</b> Recent vision-language model (VLM)-based approaches have achieved impressive results on SVG generation. However, because they generate only text and lack visual signals during decoding, they often struggle with complex semantics and fail to produce visually appealing or geometrically coherent SVGs. We introduce DuetSVG, a unified multimodal model that jointly generates image tokens and corresponding SVG tokens in an end-to-end manner. DuetSVG is trained on both image and SVG datasets. At inference, we apply a novel test-time scaling strategy that leverages the model's native visual predictions as guidance to improve SVG decoding quality. Extensive experiments show that our method outperforms existing methods, producing visually faithful, semantically aligned, and syntactically clean SVGs across a wide range of applications.</p> - https://arxiv.org/abs/2512.10894 - Thu, 11 Dec 2025 18:23:03 +0000 + Differences That Matter: Auditing Models for Capability Gap Discovery and Rectification + https://arxiv.org/abs/2512.16921 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.16921.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Qihao Liu, Chengzhi Mao, Yaojie Liu, Alan Yuille, Wen-Sheng Chu</p><p><b>Upvotes:</b> 5</p><p><b>Summary:</b> Conventional evaluation methods for multimodal LLMs (MLLMs) lack interpretability and are often insufficient to fully disclose significant capability gaps across models. To address this, we introduce AuditDM, an automated framework that actively discovers and rectifies MLLM failure modes by auditing their divergence. AuditDM fine-tunes an MLLM as an auditor via reinforcement learning to generate challenging questions and counterfactual images that maximize disagreement among target models. Once trained, the auditor uncovers diverse, interpretable exemplars that reveal model weaknesses and serve as annotation-free data for rectification. When applied to SoTA models like Gemma-3 and PaliGemma-2, AuditDM discovers more than 20 distinct failure types. Fine-tuning on these discoveries consistently improves all models across 16 benchmarks, and enables a 3B model to surpass its 28B counterpart. Our results suggest that as data scaling hits diminishing returns, targeted model auditing offers an effective path to model diagnosis and improvement.</p> + https://arxiv.org/abs/2512.16921 + Thu, 18 Dec 2025 18:59:57 +0000 - Stronger Normalization-Free Transformers - https://arxiv.org/abs/2512.10938 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.10938.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Mingzhi Chen, Taiming Lu, Jiachen Zhu, Mingjie Sun, Zhuang Liu</p><p><b>Upvotes:</b> 15</p><p><b>Summary:</b> Although normalization layers have long been viewed as indispensable components of deep learning architectures, the recent introduction of Dynamic Tanh (DyT) has demonstrated that alternatives are possible. The point-wise function DyT constrains extreme values for stable convergence and reaches normalization-level performance; this work seeks further for function designs that can surpass it. We first study how the intrinsic properties of point-wise functions influence training and performance. Building on these findings, we conduct a large-scale search for a more effective function design. Through this exploration, we introduce Derf(x) = erf(αx + s), where erf(x) is the rescaled Gaussian cumulative distribution function, and identify it as the most performant design. Derf outperforms LayerNorm, RMSNorm, and DyT across a wide range of domains, including vision (image recognition and generation), speech representation, and DNA sequence modeling. Our findings suggest that the performance gains of Derf largely stem from its improved generalization rather than stronger fitting capacity. Its simplicity and stronger performance make Derf a practical choice for normalization-free Transformer architectures.</p> - https://arxiv.org/abs/2512.10938 - Thu, 11 Dec 2025 18:58:49 +0000 + EasyV2V: A High-quality Instruction-based Video Editing Framework + https://arxiv.org/abs/2512.16920 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.16920.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jinjie Mai, Chaoyang Wang, Guocheng Gordon Qian, Willi Menapace, Sergey Tulyakov, Bernard Ghanem, Peter Wonka, Ashkan Mirzaei</p><p><b>Upvotes:</b> 12</p><p><b>Summary:</b> While image editing has advanced rapidly, video editing remains less explored, facing challenges in consistency, control, and generalization. We study the design space of data, architecture, and control, and introduce EasyV2V, a simple and effective framework for instruction-based video editing. On the data side, we compose existing experts with fast inverses to build diverse video pairs, lift image edit pairs into videos via single-frame supervision and pseudo pairs with shared affine motion, mine dense-captioned clips for video pairs, and add transition supervision to teach how edits unfold. On the model side, we observe that pretrained text-to-video models possess editing capability, motivating a simplified design. Simple sequence concatenation for conditioning with light LoRA fine-tuning suffices to train a strong model. For control, we unify spatiotemporal control via a single mask mechanism and support optional reference images. Overall, EasyV2V works with flexible inputs, e.g., video+text, video+mask+text, video+mask+reference+text, and achieves state-of-the-art video editing results, surpassing concurrent and commercial systems. Project page: https://snap-research.github.io/easyv2v/</p> + https://arxiv.org/abs/2512.16920 + Thu, 18 Dec 2025 18:59:57 +0000 - Are We Ready for RL in Text-to-3D Generation? A Progressive Investigation - https://arxiv.org/abs/2512.10949 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.10949.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yiwen Tang, Zoey Guo, Kaixin Zhu, Ray Zhang, Qizhi Chen, Dongzhi Jiang, Junli Liu, Bohan Zeng, Haoming Song, Delin Qu, Tianyi Bai, Dan Xu, Wentao Zhang, Bin Zhao</p><p><b>Upvotes:</b> 40</p><p><b>Summary:</b> Reinforcement learning (RL), earlier proven to be effective in large language and multi-modal models, has been successfully extended to enhance 2D image generation recently. However, applying RL to 3D generation remains largely unexplored due to the higher spatial complexity of 3D objects, which require globally consistent geometry and fine-grained local textures. This makes 3D generation significantly sensitive to reward designs and RL algorithms. To address these challenges, we conduct the first systematic study of RL for text-to-3D autoregressive generation across several dimensions. (1) Reward designs: We evaluate reward dimensions and model choices, showing that alignment with human preference is crucial, and that general multi-modal models provide robust signal for 3D attributes. (2) RL algorithms: We study GRPO variants, highlighting the effectiveness of token-level optimization, and further investigate the scaling of training data and iterations. (3) Text-to-3D Benchmarks: Since existing benchmarks fail to measure implicit reasoning abilities in 3D generation models, we introduce MME-3DR. (4) Advanced RL paradigms: Motivated by the natural hierarchy of 3D generation, we propose Hi-GRPO, which optimizes the global-to-local hierarchical 3D generation through dedicated reward ensembles. Based on these insights, we develop AR3D-R1, the first RL-enhanced text-to-3D model, expert from coarse shape to texture refinement. We hope this study provides insights into RL-driven reasoning for 3D generation. Code is released at https://github.com/Ivan-Tang-3D/3DGen-R1.</p> - https://arxiv.org/abs/2512.10949 - Thu, 11 Dec 2025 18:59:52 +0000 + Next-Embedding Prediction Makes Strong Vision Learners + https://arxiv.org/abs/2512.16922 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.16922.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Sihan Xu, Ziqiao Ma, Wenhao Chai, Xuweiyi Chen, Weiyang Jin, Joyce Chai, Saining Xie, Stella X. Yu</p><p><b>Upvotes:</b> 59</p><p><b>Summary:</b> Inspired by the success of generative pretraining in natural language, we ask whether the same principles can yield strong self-supervised visual learners. Instead of training models to output features for downstream use, we train them to generate embeddings to perform predictive tasks directly. This work explores such a shift from learning representations to learning models. Specifically, models learn to predict future patch embeddings conditioned on past ones, using causal masking and stop gradient, which we refer to as Next-Embedding Predictive Autoregression (NEPA). We demonstrate that a simple Transformer pretrained on ImageNet-1k with next embedding prediction as its sole learning objective is effective - no pixel reconstruction, discrete tokens, contrastive loss, or task-specific heads. This formulation retains architectural simplicity and scalability, without requiring additional design complexity. NEPA achieves strong results across tasks, attaining 83.8% and 85.3% top-1 accuracy on ImageNet-1K with ViT-B and ViT-L backbones after fine-tuning, and transferring effectively to semantic segmentation on ADE20K. We believe generative pretraining from embeddings provides a simple, scalable, and potentially modality-agnostic alternative to visual self-supervised learning.</p> + https://arxiv.org/abs/2512.16922 + Thu, 18 Dec 2025 18:59:58 +0000 - Omni-Attribute: Open-vocabulary Attribute Encoder for Visual Concept Personalization - https://arxiv.org/abs/2512.10955 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.10955.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Tsai-Shien Chen, Aliaksandr Siarohin, Guocheng Gordon Qian, Kuan-Chieh Jackson Wang, Egor Nemchinov, Moayed Haji-Ali, Riza Alp Guler, Willi Menapace, Ivan Skorokhodov, Anil Kag, Jun-Yan Zhu, Sergey Tulyakov</p><p><b>Upvotes:</b> 4</p><p><b>Summary:</b> Visual concept personalization aims to transfer only specific image attributes, such as identity, expression, lighting, and style, into unseen contexts. However, existing methods rely on holistic embeddings from general-purpose image encoders, which entangle multiple visual factors and make it difficult to isolate a single attribute. This often leads to information leakage and incoherent synthesis. To address this limitation, we introduce Omni-Attribute, the first open-vocabulary image attribute encoder designed to learn high-fidelity, attribute-specific representations. Our approach jointly designs the data and model: (i) we curate semantically linked image pairs annotated with positive and negative attributes to explicitly teach the encoder what to preserve or suppress; and (ii) we adopt a dual-objective training paradigm that balances generative fidelity with contrastive disentanglement. The resulting embeddings prove effective for open-vocabulary attribute retrieval, personalization, and compositional generation, achieving state-of-the-art performance across multiple benchmarks.</p> - https://arxiv.org/abs/2512.10955 - Thu, 11 Dec 2025 18:59:56 +0000 + The World is Your Canvas: Painting Promptable Events with Reference Images, Trajectories, and Text + https://arxiv.org/abs/2512.16924 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.16924.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Hanlin Wang, Hao Ouyang, Qiuyu Wang, Yue Yu, Yihao Meng, Wen Wang, Ka Leong Cheng, Shuailei Ma, Qingyan Bai, Yixuan Li, Cheng Chen, Yanhong Zeng, Xing Zhu, Yujun Shen, Qifeng Chen</p><p><b>Upvotes:</b> 21</p><p><b>Summary:</b> We present WorldCanvas, a framework for promptable world events that enables rich, user-directed simulation by combining text, trajectories, and reference images. Unlike text-only approaches and existing trajectory-controlled image-to-video methods, our multimodal approach combines trajectories -- encoding motion, timing, and visibility -- with natural language for semantic intent and reference images for visual grounding of object identity, enabling the generation of coherent, controllable events that include multi-agent interactions, object entry/exit, reference-guided appearance and counterintuitive events. The resulting videos demonstrate not only temporal coherence but also emergent consistency, preserving object identity and scene despite temporary disappearance. By supporting expressive world events generation, WorldCanvas advances world models from passive predictors to interactive, user-shaped simulators. Our project page is available at: https://worldcanvas.github.io/.</p> + https://arxiv.org/abs/2512.16924 + Thu, 18 Dec 2025 18:59:59 +0000 - StereoSpace: Depth-Free Synthesis of Stereo Geometry via End-to-End Diffusion in a Canonical Space - https://arxiv.org/abs/2512.10959 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.10959.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Tjark Behrens, Anton Obukhov, Bingxin Ke, Fabio Tosi, Matteo Poggi, Konrad Schindler</p><p><b>Upvotes:</b> 10</p><p><b>Summary:</b> We introduce StereoSpace, a diffusion-based framework for monocular-to-stereo synthesis that models geometry purely through viewpoint conditioning, without explicit depth or warping. A canonical rectified space and the conditioning guide the generator to infer correspondences and fill disocclusions end-to-end. To ensure fair and leakage-free evaluation, we introduce an end-to-end protocol that excludes any ground truth or proxy geometry estimates at test time. The protocol emphasizes metrics reflecting downstream relevance: iSQoE for perceptual comfort and MEt3R for geometric consistency. StereoSpace surpasses other methods from the warp & inpaint, latent-warping, and warped-conditioning categories, achieving sharp parallax and strong robustness on layered and non-Lambertian scenes. This establishes viewpoint-conditioned diffusion as a scalable, depth-free solution for stereo generation.</p> - https://arxiv.org/abs/2512.10959 - Thu, 11 Dec 2025 18:59:59 +0000 + Generative Refocusing: Flexible Defocus Control from a Single Image + https://arxiv.org/abs/2512.16923 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2512.16923.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Chun-Wei Tuan Mu, Jia-Bin Huang, Yu-Lun Liu</p><p><b>Upvotes:</b> 32</p><p><b>Summary:</b> Depth-of-field control is essential in photography, but getting the perfect focus often takes several tries or special equipment. Single-image refocusing is still difficult. It involves recovering sharp content and creating realistic bokeh. Current methods have significant drawbacks. They need all-in-focus inputs, depend on synthetic data from simulators, and have limited control over aperture. We introduce Generative Refocusing, a two-step process that uses DeblurNet to recover all-in-focus images from various inputs and BokehNet for creating controllable bokeh. Our main innovation is semi-supervised training. This method combines synthetic paired data with unpaired real bokeh images, using EXIF metadata to capture real optical characteristics beyond what simulators can provide. Our experiments show we achieve top performance in defocus deblurring, bokeh synthesis, and refocusing benchmarks. Additionally, our Generative Refocusing allows text-guided adjustments and custom aperture shapes.</p> + https://arxiv.org/abs/2512.16923 + Thu, 18 Dec 2025 18:59:59 +0000