diff --git a/feed.xml b/feed.xml index 1742553..63fadfb 100644 --- a/feed.xml +++ b/feed.xml @@ -7,174 +7,104 @@ http://www.rssboard.org/rss-specification python-feedgen en - Wed, 14 May 2025 00:02:29 +0000 + Thu, 15 May 2025 00:02:42 +0000 - Position: AI Competitions Provide the Gold Standard for Empirical Rigor in GenAI Evaluation - https://arxiv.org/abs/2505.00612 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.00612.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> D. Sculley, Will Cukierski, Phil Culliton, Sohier Dane, Maggie Demkin, Ryan Holbrook, Addison Howard, Paul Mooney, Walter Reade, Megan Risdal, Nate Keating</p><p><b>Upvotes:</b> 5</p><p><b>Summary:</b> In this position paper, we observe that empirical evaluation in Generative AI is at a crisis point since traditional ML evaluation and benchmarking strategies are insufficient to meet the needs of evaluating modern GenAI models and systems. There are many reasons for this, including the fact that these models typically have nearly unbounded input and output spaces, typically do not have a well defined ground truth target, and typically exhibit strong feedback loops and prediction dependence based on context of previous model outputs. On top of these critical issues, we argue that the problems of {\em leakage} and {\em contamination} are in fact the most important and difficult issues to address for GenAI evaluations. Interestingly, the field of AI Competitions has developed effective measures and practices to combat leakage for the purpose of counteracting cheating by bad actors within a competition setting. This makes AI Competitions an especially valuable (but underutilized) resource. Now is time for the field to view AI Competitions as the gold standard for empirical rigor in GenAI evaluation, and to harness and harvest their results with according value.</p> - https://arxiv.org/abs/2505.00612 - Thu, 01 May 2025 15:43:51 +0000 + Advancing Arabic Reverse Dictionary Systems: A Transformer-Based Approach with Dataset Construction Guidelines + https://arxiv.org/abs/2504.21475 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2504.21475.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Serry Sibaee, Samar Ahmed, Abdullah Al Harbi, Omer Nacar, Adel Ammar, Yasser Habashi, Wadii Boulila</p><p><b>Upvotes:</b> 2</p><p><b>Summary:</b> This study addresses the critical gap in Arabic natural language processing by developing an effective Arabic Reverse Dictionary (RD) system that enables users to find words based on their descriptions or meanings. We present a novel transformer-based approach with a semi-encoder neural network architecture featuring geometrically decreasing layers that achieves state-of-the-art results for Arabic RD tasks. Our methodology incorporates a comprehensive dataset construction process and establishes formal quality standards for Arabic lexicographic definitions. Experiments with various pre-trained models demonstrate that Arabic-specific models significantly outperform general multilingual embeddings, with ARBERTv2 achieving the best ranking score (0.0644). Additionally, we provide a formal abstraction of the reverse dictionary task that enhances theoretical understanding and develop a modular, extensible Python library (RDTL) with configurable training pipelines. Our analysis of dataset quality reveals important insights for improving Arabic definition construction, leading to eight specific standards for building high-quality reverse dictionary resources. This work contributes significantly to Arabic computational linguistics and provides valuable tools for language learning, academic writing, and professional communication in Arabic.</p> + https://arxiv.org/abs/2504.21475 + Wed, 30 Apr 2025 09:56:36 +0000 - WebGen-Bench: Evaluating LLMs on Generating Interactive and Functional Websites from Scratch - https://arxiv.org/abs/2505.03733 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.03733.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zimu Lu, Yunqiao Yang, Houxing Ren, Haotian Hou, Han Xiao, Ke Wang, Weikang Shi, Aojun Zhou, Mingjie Zhan, Hongsheng Li</p><p><b>Upvotes:</b> 15</p><p><b>Summary:</b> LLM-based agents have demonstrated great potential in generating and managing code within complex codebases. In this paper, we introduce WebGen-Bench, a novel benchmark designed to measure an LLM-based agent's ability to create multi-file website codebases from scratch. It contains diverse instructions for website generation, created through the combined efforts of human annotators and GPT-4o. These instructions span three major categories and thirteen minor categories, encompassing nearly all important types of web applications. To assess the quality of the generated websites, we use GPT-4o to generate test cases targeting each functionality described in the instructions, and then manually filter, adjust, and organize them to ensure accuracy, resulting in 647 test cases. Each test case specifies an operation to be performed on the website and the expected result after the operation. To automate testing and improve reproducibility, we employ a powerful web-navigation agent to execute tests on the generated websites and determine whether the observed responses align with the expected results. We evaluate three high-performance code-agent frameworks, Bolt.diy, OpenHands, and Aider, using multiple proprietary and open-source LLMs as engines. The best-performing combination, Bolt.diy powered by DeepSeek-R1, achieves only 27.8\% accuracy on the test cases, highlighting the challenging nature of our benchmark. Additionally, we construct WebGen-Instruct, a training set consisting of 6,667 website-generation instructions. Training Qwen2.5-Coder-32B-Instruct on Bolt.diy trajectories generated from a subset of this training set achieves an accuracy of 38.2\%, surpassing the performance of the best proprietary model.</p> - https://arxiv.org/abs/2505.03733 - Tue, 06 May 2025 17:59:15 +0000 + Bring Reason to Vision: Understanding Perception and Reasoning through Model Merging + https://arxiv.org/abs/2505.05464 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.05464.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Shiqi Chen, Jinghan Zhang, Tongyao Zhu, Wei Liu, Siyang Gao, Miao Xiong, Manling Li, Junxian He</p><p><b>Upvotes:</b> 5</p><p><b>Summary:</b> Vision-Language Models (VLMs) combine visual perception with the general capabilities, such as reasoning, of Large Language Models (LLMs). However, the mechanisms by which these two abilities can be combined and contribute remain poorly understood. In this work, we explore to compose perception and reasoning through model merging that connects parameters of different models. Unlike previous works that often focus on merging models of the same kind, we propose merging models across modalities, enabling the incorporation of the reasoning capabilities of LLMs into VLMs. Through extensive experiments, we demonstrate that model merging offers a successful pathway to transfer reasoning abilities from LLMs to VLMs in a training-free manner. Moreover, we utilize the merged models to understand the internal mechanism of perception and reasoning and how merging affects it. We find that perception capabilities are predominantly encoded in the early layers of the model, whereas reasoning is largely facilitated by the middle-to-late layers. After merging, we observe that all layers begin to contribute to reasoning, whereas the distribution of perception abilities across layers remains largely unchanged. These observations shed light on the potential of model merging as a tool for multimodal integration and interpretation.</p> + https://arxiv.org/abs/2505.05464 + Thu, 08 May 2025 17:56:23 +0000 - LLAMAPIE: Proactive In-Ear Conversation Assistants - https://arxiv.org/abs/2505.04066 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.04066.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Tuochao Chen, Nicholas Batchelder, Alisa Liu, Noah Smith, Shyamnath Gollakota</p><p><b>Upvotes:</b> 1</p><p><b>Summary:</b> We introduce LlamaPIE, the first real-time proactive assistant designed to enhance human conversations through discreet, concise guidance delivered via hearable devices. Unlike traditional language models that require explicit user invocation, this assistant operates in the background, anticipating user needs without interrupting conversations. We address several challenges, including determining when to respond, crafting concise responses that enhance conversations, leveraging knowledge of the user for context-aware assistance, and real-time, on-device processing. To achieve this, we construct a semi-synthetic dialogue dataset and propose a two-model pipeline: a small model that decides when to respond and a larger model that generates the response. We evaluate our approach on real-world datasets, demonstrating its effectiveness in providing helpful, unobtrusive assistance. User studies with our assistant, implemented on Apple Silicon M2 hardware, show a strong preference for the proactive assistant over both a baseline with no assistance and a reactive model, highlighting the potential of LlamaPie to enhance live conversations.</p> - https://arxiv.org/abs/2505.04066 - Wed, 07 May 2025 02:08:56 +0000 + Measuring General Intelligence with Generated Games + https://arxiv.org/abs/2505.07215 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.07215.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Vivek Verma, David Huang, William Chen, Dan Klein, Nicholas Tomlin</p><p><b>Upvotes:</b> 5</p><p><b>Summary:</b> We present gg-bench, a collection of game environments designed to evaluate general reasoning capabilities in language models. Unlike most static benchmarks, gg-bench is a data generating process where new evaluation instances can be generated at will. In particular, gg-bench is synthetically generated by (1) using a large language model (LLM) to generate natural language descriptions of novel games, (2) using the LLM to implement each game in code as a Gym environment, and (3) training reinforcement learning (RL) agents via self-play on the generated games. We evaluate language models by their winrate against these RL agents by prompting models with the game description, current board state, and a list of valid moves, after which models output the moves they wish to take. gg-bench is challenging: state-of-the-art LLMs such as GPT-4o and Claude 3.7 Sonnet achieve winrates of 7-9% on gg-bench using in-context learning, while reasoning models such as o1, o3-mini and DeepSeek-R1 achieve average winrates of 31-36%. We release the generated games, data generation process, and evaluation code in order to support future modeling work and expansion of our benchmark.</p> + https://arxiv.org/abs/2505.07215 + Mon, 12 May 2025 04:01:03 +0000 - Physics-Assisted and Topology-Informed Deep Learning for Weather Prediction - https://arxiv.org/abs/2505.04918 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.04918.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jiaqi Zheng, Qing Ling, Yerong Feng</p><p><b>Upvotes:</b> 2</p><p><b>Summary:</b> Although deep learning models have demonstrated remarkable potential in weather prediction, most of them overlook either the physics of the underlying weather evolution or the topology of the Earth's surface. In light of these disadvantages, we develop PASSAT, a novel Physics-ASSisted And Topology-informed deep learning model for weather prediction. PASSAT attributes the weather evolution to two key factors: (i) the advection process that can be characterized by the advection equation and the Navier-Stokes equation; (ii) the Earth-atmosphere interaction that is difficult to both model and calculate. PASSAT also takes the topology of the Earth's surface into consideration, other than simply treating it as a plane. With these considerations, PASSAT numerically solves the advection equation and the Navier-Stokes equation on the spherical manifold, utilizes a spherical graph neural network to capture the Earth-atmosphere interaction, and generates the initial velocity fields that are critical to solving the advection equation from the same spherical graph neural network. In the 5.625^circ-resolution ERA5 data set, PASSAT outperforms both the state-of-the-art deep learning-based weather prediction models and the operational numerical weather prediction model IFS T42. Code and checkpoint are available at https://github.com/Yumenomae/PASSAT_5p625.</p> - https://arxiv.org/abs/2505.04918 - Thu, 08 May 2025 03:25:55 +0000 + ViMRHP: A Vietnamese Benchmark Dataset for Multimodal Review Helpfulness Prediction via Human-AI Collaborative Annotation + https://arxiv.org/abs/2505.07416 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.07416.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Truc Mai-Thanh Nguyen, Dat Minh Nguyen, Son T. Luu, Kiet Van Nguyen</p><p><b>Upvotes:</b> 1</p><p><b>Summary:</b> Multimodal Review Helpfulness Prediction (MRHP) is an essential task in recommender systems, particularly in E-commerce platforms. Determining the helpfulness of user-generated reviews enhances user experience and improves consumer decision-making. However, existing datasets focus predominantly on English and Indonesian, resulting in a lack of linguistic diversity, especially for low-resource languages such as Vietnamese. In this paper, we introduce ViMRHP (Vietnamese Multimodal Review Helpfulness Prediction), a large-scale benchmark dataset for MRHP task in Vietnamese. This dataset covers four domains, including 2K products with 46K reviews. Meanwhile, a large-scale dataset requires considerable time and cost. To optimize the annotation process, we leverage AI to assist annotators in constructing the ViMRHP dataset. With AI assistance, annotation time is reduced (90 to 120 seconds per task down to 20 to 40 seconds per task) while maintaining data quality and lowering overall costs by approximately 65%. However, AI-generated annotations still have limitations in complex annotation tasks, which we further examine through a detailed performance analysis. In our experiment on ViMRHP, we evaluate baseline models on human-verified and AI-generated annotations to assess their quality differences. The ViMRHP dataset is publicly available at https://github.com/trng28/ViMRHP</p> + https://arxiv.org/abs/2505.07416 + Mon, 12 May 2025 10:11:28 +0000 - Document Attribution: Examining Citation Relationships using Large Language Models - https://arxiv.org/abs/2505.06324 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.06324.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Vipula Rawte, Ryan A. Rossi, Franck Dernoncourt, Nedim Lipka</p><p><b>Upvotes:</b> 3</p><p><b>Summary:</b> As Large Language Models (LLMs) are increasingly applied to document-based tasks - such as document summarization, question answering, and information extraction - where user requirements focus on retrieving information from provided documents rather than relying on the model's parametric knowledge, ensuring the trustworthiness and interpretability of these systems has become a critical concern. A central approach to addressing this challenge is attribution, which involves tracing the generated outputs back to their source documents. However, since LLMs can produce inaccurate or imprecise responses, it is crucial to assess the reliability of these citations. To tackle this, our work proposes two techniques. (1) A zero-shot approach that frames attribution as a straightforward textual entailment task. Our method using flan-ul2 demonstrates an improvement of 0.27% and 2.4% over the best baseline of ID and OOD sets of AttributionBench, respectively. (2) We also explore the role of the attention mechanism in enhancing the attribution process. Using a smaller LLM, flan-t5-small, the F1 scores outperform the baseline across almost all layers except layer 4 and layers 8 through 11.</p> - https://arxiv.org/abs/2505.06324 - Fri, 09 May 2025 04:40:11 +0000 + A Multi-Dimensional Constraint Framework for Evaluating and Improving Instruction Following in Large Language Models + https://arxiv.org/abs/2505.07591 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.07591.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Junjie Ye, Caishuang Huang, Zhuohan Chen, Wenjie Fu, Chenyuan Yang, Leyi Yang, Yilong Wu, Peng Wang, Meng Zhou, Xiaolong Yang, Tao Gui, Qi Zhang, Zhongchao Shi, Jianping Fan, Xuanjing Huang</p><p><b>Upvotes:</b> 7</p><p><b>Summary:</b> Instruction following evaluates large language models (LLMs) on their ability to generate outputs that adhere to user-defined constraints. However, existing benchmarks often rely on templated constraint prompts, which lack the diversity of real-world usage and limit fine-grained performance assessment. To fill this gap, we propose a multi-dimensional constraint framework encompassing three constraint patterns, four constraint categories, and four difficulty levels. Building on this framework, we develop an automated instruction generation pipeline that performs constraint expansion, conflict detection, and instruction rewriting, yielding 1,200 code-verifiable instruction-following test samples. We evaluate 19 LLMs across seven model families and uncover substantial variation in performance across constraint forms. For instance, average performance drops from 77.67% at Level I to 32.96% at Level IV. Furthermore, we demonstrate the utility of our approach by using it to generate data for reinforcement learning, achieving substantial gains in instruction following without degrading general performance. In-depth analysis indicates that these gains stem primarily from modifications in the model's attention modules parameters, which enhance constraint recognition and adherence. Code and data are available in https://github.com/Junjie-Ye/MulDimIF.</p> + https://arxiv.org/abs/2505.07591 + Mon, 12 May 2025 14:16:55 +0000 - MonetGPT: Solving Puzzles Enhances MLLMs' Image Retouching Skills - https://arxiv.org/abs/2505.06176 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.06176.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Niladri Shekhar Dutt, Duygu Ceylan, Niloy J. Mitra</p><p><b>Upvotes:</b> 7</p><p><b>Summary:</b> Retouching is an essential task in post-manipulation of raw photographs. Generative editing, guided by text or strokes, provides a new tool accessible to users but can easily change the identity of the original objects in unacceptable and unpredictable ways. In contrast, although traditional procedural edits, as commonly supported by photoediting tools (e.g., Gimp, Lightroom), are conservative, they are still preferred by professionals. Unfortunately, professional quality retouching involves many individual procedural editing operations that is challenging to plan for most novices. In this paper, we ask if a multimodal large language model (MLLM) can be taught to critique raw photographs, suggest suitable remedies, and finally realize them with a given set of pre-authored procedural image operations. We demonstrate that MLLMs can be first made aware of the underlying image processing operations, by training them to solve specially designed visual puzzles. Subsequently, such an operation-aware MLLM can both plan and propose edit sequences. To facilitate training, given a set of expert-edited photos, we synthesize a reasoning dataset by procedurally manipulating the expert edits and then grounding a pretrained LLM on the visual adjustments, to synthesize reasoning for finetuning. The proposed retouching operations are, by construction, understandable by the users, preserve object details and resolution, and can be optionally overridden. We evaluate our setup on a variety of test examples and show advantages, in terms of explainability and identity preservation, over existing generative and other procedural alternatives. Code, data, models, and supplementary results can be found via our project website at https://monetgpt.github.io.</p> - https://arxiv.org/abs/2505.06176 - Fri, 09 May 2025 16:38:27 +0000 + MiniMax-Speech: Intrinsic Zero-Shot Text-to-Speech with a Learnable Speaker Encoder + https://arxiv.org/abs/2505.07916 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.07916.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Bowen Zhang, Congchao Guo, Geng Yang, Hang Yu, Haozhe Zhang, Heidi Lei, Jialong Mai, Junjie Yan, Kaiyue Yang, Mingqi Yang, Peikai Huang, Ruiyang Jin, Sitan Jiang, Weihua Cheng, Yawei Li, Yichen Xiao, Yiying Zhou, Yongmao Zhang, Yuan Lu, Yucen He</p><p><b>Upvotes:</b> 81</p><p><b>Summary:</b> We introduce MiniMax-Speech, an autoregressive Transformer-based Text-to-Speech (TTS) model that generates high-quality speech. A key innovation is our learnable speaker encoder, which extracts timbre features from a reference audio without requiring its transcription. This enables MiniMax-Speech to produce highly expressive speech with timbre consistent with the reference in a zero-shot manner, while also supporting one-shot voice cloning with exceptionally high similarity to the reference voice. In addition, the overall quality of the synthesized audio is enhanced through the proposed Flow-VAE. Our model supports 32 languages and demonstrates excellent performance across multiple objective and subjective evaluations metrics. Notably, it achieves state-of-the-art (SOTA) results on objective voice cloning metrics (Word Error Rate and Speaker Similarity) and has secured the top position on the public TTS Arena leaderboard. Another key strength of MiniMax-Speech, granted by the robust and disentangled representations from the speaker encoder, is its extensibility without modifying the base model, enabling various applications such as: arbitrary voice emotion control via LoRA; text to voice (T2V) by synthesizing timbre features directly from text description; and professional voice cloning (PVC) by fine-tuning timbre features with additional data. We encourage readers to visit https://minimax-ai.github.io/tts_tech_report for more examples.</p> + https://arxiv.org/abs/2505.07916 + Mon, 12 May 2025 14:25:20 +0000 - REFINE-AF: A Task-Agnostic Framework to Align Language Models via Self-Generated Instructions using Reinforcement Learning from Automated Feedback - https://arxiv.org/abs/2505.06548 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.06548.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Aniruddha Roy, Pretam Ray, Abhilash Nandy, Somak Aditya, Pawan Goyal</p><p><b>Upvotes:</b> 26</p><p><b>Summary:</b> Instruction-based Large Language Models (LLMs) have proven effective in numerous few-shot or zero-shot Natural Language Processing (NLP) tasks. However, creating human-annotated instruction data is time-consuming, expensive, and often limited in quantity and task diversity. Previous research endeavors have attempted to address this challenge by proposing frameworks capable of generating instructions in a semi-automated and task-agnostic manner directly from the model itself. Many of these efforts have relied on large API-only parameter-based models such as GPT-3.5 (175B), which are expensive, and subject to limits on a number of queries. This paper explores the performance of three open-source small LLMs such as LLaMA 2-7B, LLama 2-13B, and Mistral 7B, using a semi-automated framework, thereby reducing human intervention, effort, and cost required to generate an instruction dataset for fine-tuning LLMs. Furthermore, we demonstrate that incorporating a Reinforcement Learning (RL) based training algorithm into this LLMs-based framework leads to further enhancements. Our evaluation of the dataset reveals that these RL-based frameworks achieve a substantial improvements in 63-66% of the tasks compared to previous approaches.</p> - https://arxiv.org/abs/2505.06548 - Sat, 10 May 2025 07:23:19 +0000 + Fast Text-to-Audio Generation with Adversarial Post-Training + https://arxiv.org/abs/2505.08175 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.08175.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zachary Novack, Zach Evans, Zack Zukowski, Josiah Taylor, CJ Carr, Julian Parker, Adnan Al-Sinan, Gian Marco Iodice, Julian McAuley, Taylor Berg-Kirkpatrick, Jordi Pons</p><p><b>Upvotes:</b> 6</p><p><b>Summary:</b> Text-to-audio systems, while increasingly performant, are slow at inference time, thus making their latency unpractical for many creative applications. We present Adversarial Relativistic-Contrastive (ARC) post-training, the first adversarial acceleration algorithm for diffusion/flow models not based on distillation. While past adversarial post-training methods have struggled to compare against their expensive distillation counterparts, ARC post-training is a simple procedure that (1) extends a recent relativistic adversarial formulation to diffusion/flow post-training and (2) combines it with a novel contrastive discriminator objective to encourage better prompt adherence. We pair ARC post-training with a number optimizations to Stable Audio Open and build a model capable of generating approx12s of 44.1kHz stereo audio in approx75ms on an H100, and approx7s on a mobile edge-device, the fastest text-to-audio model to our knowledge.</p> + https://arxiv.org/abs/2505.08175 + Tue, 13 May 2025 02:25:47 +0000 - Seed1.5-VL Technical Report - https://arxiv.org/abs/2505.07062 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.07062.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Dong Guo, Faming Wu, Feida Zhu, Fuxing Leng, Guang Shi, Haobin Chen, Haoqi Fan, Jian Wang, Jianyu Jiang, Jiawei Wang, Jingji Chen, Jingjia Huang, Kang Lei, Liping Yuan, Lishu Luo, Pengfei Liu, Qinghao Ye, Rui Qian, Shen Yan, Shixiong Zhao, Shuai Peng, Shuangye Li, Sihang Yuan, Sijin Wu, Tianheng Cheng, Weiwei Liu, Wenqian Wang, Xianhan Zeng, Xiao Liu, Xiaobo Qin, Xiaohan Ding, Xiaojun Xiao, Xiaoying Zhang, Xuanwei Zhang, Xuehan Xiong, Yanghua Peng, Yangrui Chen, Yanwei Li, Yanxu Hu, Yi Lin, Yiyuan Hu, Yiyuan Zhang, Youbin Wu, Yu Li, Yudong Liu, Yue Ling, Yujia Qin, Zanbo Wang, Zhiwu He, Aoxue Zhang, Bairen Yi, Bencheng Liao, Can Huang, Can Zhang, Chaorui Deng, Chaoyi Deng, Cheng Lin, Cheng Yuan, Chenggang Li, Chenhui Gou, Chenwei Lou, Chengzhi Wei, Chundian Liu, Chunyuan Li, Deyao Zhu, Donghong Zhong, Feng Li, Feng Zhang, Gang Wu, Guodong Li, Guohong Xiao, Haibin Lin, Haihua Yang, Haoming Wang, Heng Ji, Hongxiang Hao, Hui Shen, Huixia Li, Jiahao Li, Jialong Wu, Jianhua Zhu, Jianpeng Jiao, Jiashi Feng, Jiaze Chen, Jianhui Duan, Jihao Liu, Jin Zeng, Jingqun Tang, Jingyu Sun, Joya Chen, Jun Long, Junda Feng, Junfeng Zhan, Junjie Fang, Junting Lu, Kai Hua, Kai Liu, Kai Shen, Kaiyuan Zhang, Ke Shen, Ke Wang, Keyu Pan, Kun Zhang, Kunchang Li, Lanxin Li, Lei Li, Lei Shi, Li Han, Liang Xiang, Liangqiang Chen, Lin Chen, Lin Li, Lin Yan, Liying Chi, Longxiang Liu, Mengfei Du, Mingxuan Wang, Ningxin Pan, Peibin Chen, Pengfei Chen, Pengfei Wu, Qingqing Yuan, Qingyao Shuai, Qiuyan Tao, Renjie Zheng, Renrui Zhang, Ru Zhang, Rui Wang, Rui Yang, Rui Zhao, Shaoqiang Xu, Shihao Liang, Shipeng Yan, Shu Zhong, Shuaishuai Cao, Shuangzhi Wu, Shufan Liu, Shuhan Chang, Songhua Cai, Tenglong Ao, Tianhao Yang, Tingting Zhang, Wanjun Zhong, Wei Jia, Wei Weng, Weihao Yu, Wenhao Huang, Wenjia Zhu, Wenli Yang, Wenzhi Wang, Xiang Long, XiangRui Yin, Xiao Li, Xiaolei Zhu, Xiaoying Jia, Xijin Zhang, Xin Liu, Xinchen Zhang, Xinyu Yang, Xiongcai Luo, Xiuli Chen, Xuantong Zhong, Xuefeng Xiao, Xujing Li, Yan Wu, Yawei Wen, Yifan Du, Yihao Zhang, Yining Ye, Yonghui Wu, Yu Liu, Yu Yue, Yufeng Zhou, Yufeng Yuan, Yuhang Xu, Yuhong Yang, Yun Zhang, Yunhao Fang, Yuntao Li, Yurui Ren, Yuwen Xiong, Zehua Hong, Zehua Wang, Zewei Sun, Zeyu Wang, Zhao Cai, Zhaoyue Zha, Zhecheng An, Zhehui Zhao, Zhengzhuo Xu, Zhipeng Chen, Zhiyong Wu, Zhuofan Zheng, Zihao Wang, Zilong Huang, Ziyu Zhu, Zuquan Song</p><p><b>Upvotes:</b> 83</p><p><b>Summary:</b> We present Seed1.5-VL, a vision-language foundation model designed to advance general-purpose multimodal understanding and reasoning. Seed1.5-VL is composed with a 532M-parameter vision encoder and a Mixture-of-Experts (MoE) LLM of 20B active parameters. Despite its relatively compact architecture, it delivers strong performance across a wide spectrum of public VLM benchmarks and internal evaluation suites, achieving the state-of-the-art performance on 38 out of 60 public benchmarks. Moreover, in agent-centric tasks such as GUI control and gameplay, Seed1.5-VL outperforms leading multimodal systems, including OpenAI CUA and Claude 3.7. Beyond visual and video understanding, it also demonstrates strong reasoning abilities, making it particularly effective for multimodal reasoning challenges such as visual puzzles. We believe these capabilities will empower broader applications across diverse tasks. In this report, we mainly provide a comprehensive review of our experiences in building Seed1.5-VL across model design, data construction, and training at various stages, hoping that this report can inspire further research. Seed1.5-VL is now accessible at https://www.volcengine.com/ (Volcano Engine Model ID: doubao-1-5-thinking-vision-pro-250428)</p> - https://arxiv.org/abs/2505.07062 - Sun, 11 May 2025 17:28:30 +0000 + AM-Thinking-v1: Advancing the Frontier of Reasoning at 32B Scale + https://arxiv.org/abs/2505.08311 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.08311.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yunjie Ji, Xiaoyu Tian, Sitong Zhao, Haotian Wang, Shuaiting Chen, Yiping Peng, Han Zhao, Xiangang Li</p><p><b>Upvotes:</b> 3</p><p><b>Summary:</b> We present AM-Thinking-v1, a 32B dense language model that advances the frontier of reasoning, embodying the collaborative spirit of open-source innovation. Outperforming DeepSeek-R1 and rivaling leading Mixture-of-Experts (MoE) models like Qwen3-235B-A22B and Seed1.5-Thinking, AM-Thinking-v1 achieves impressive scores of 85.3 on AIME 2024, 74.4 on AIME 2025, and 70.3 on LiveCodeBench, showcasing state-of-the-art mathematical and coding capabilities among open-source models of similar scale. Built entirely from the open-source Qwen2.5-32B base model and publicly available queries, AM-Thinking-v1 leverages a meticulously crafted post-training pipeline - combining supervised fine-tuning and reinforcement learning - to deliver exceptional reasoning capabilities. This work demonstrates that the open-source community can achieve high performance at the 32B scale, a practical sweet spot for deployment and fine-tuning. By striking a balance between top-tier performance and real-world usability, we hope AM-Thinking-v1 inspires further collaborative efforts to harness mid-scale models, pushing reasoning boundaries while keeping accessibility at the core of innovation. We have open-sourced our model on https://huggingface.co/a-m-team/AM-Thinking-v1{Hugging Face}.</p> + https://arxiv.org/abs/2505.08311 + Tue, 13 May 2025 07:41:15 +0000 - Multi-Objective-Guided Discrete Flow Matching for Controllable Biological Sequence Design - https://arxiv.org/abs/2505.07086 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.07086.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Tong Chen, Yinuo Zhang, Sophia Tang, Pranam Chatterjee</p><p><b>Upvotes:</b> 0</p><p><b>Summary:</b> Designing biological sequences that satisfy multiple, often conflicting, functional and biophysical criteria remains a central challenge in biomolecule engineering. While discrete flow matching models have recently shown promise for efficient sampling in high-dimensional sequence spaces, existing approaches address only single objectives or require continuous embeddings that can distort discrete distributions. We present Multi-Objective-Guided Discrete Flow Matching (MOG-DFM), a general framework to steer any pretrained discrete-time flow matching generator toward Pareto-efficient trade-offs across multiple scalar objectives. At each sampling step, MOG-DFM computes a hybrid rank-directional score for candidate transitions and applies an adaptive hypercone filter to enforce consistent multi-objective progression. We also trained two unconditional discrete flow matching models, PepDFM for diverse peptide generation and EnhancerDFM for functional enhancer DNA generation, as base generation models for MOG-DFM. We demonstrate MOG-DFM's effectiveness in generating peptide binders optimized across five properties (hemolysis, non-fouling, solubility, half-life, and binding affinity), and in designing DNA sequences with specific enhancer classes and DNA shapes. In total, MOG-DFM proves to be a powerful tool for multi-property-guided biomolecule sequence design.</p> - https://arxiv.org/abs/2505.07086 - Sun, 11 May 2025 18:17:44 +0000 + Optimizing Retrieval-Augmented Generation: Analysis of Hyperparameter Impact on Performance and Efficiency + https://arxiv.org/abs/2505.08445 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.08445.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Adel Ammar, Anis Koubaa, Omer Nacar, Wadii Boulila</p><p><b>Upvotes:</b> 1</p><p><b>Summary:</b> Large language models achieve high task performance yet often hallucinate or rely on outdated knowledge. Retrieval-augmented generation (RAG) addresses these gaps by coupling generation with external search. We analyse how hyperparameters influence speed and quality in RAG systems, covering Chroma and Faiss vector stores, chunking policies, cross-encoder re-ranking, and temperature, and we evaluate six metrics: faithfulness, answer correctness, answer relevancy, context precision, context recall, and answer similarity. Chroma processes queries 13% faster, whereas Faiss yields higher retrieval precision, revealing a clear speed-accuracy trade-off. Naive fixed-length chunking with small windows and minimal overlap outperforms semantic segmentation while remaining the quickest option. Re-ranking provides modest gains in retrieval quality yet increases runtime by roughly a factor of 5, so its usefulness depends on latency constraints. These results help practitioners balance computational cost and accuracy when tuning RAG systems for transparent, up-to-date responses. Finally, we re-evaluate the top configurations with a corrective RAG workflow and show that their advantages persist when the model can iteratively request additional evidence. We obtain a near-perfect context precision (99%), which demonstrates that RAG systems can achieve extremely high retrieval accuracy with the right combination of hyperparameters, with significant implications for applications where retrieval quality directly impacts downstream task performance, such as clinical decision support in healthcare.</p> + https://arxiv.org/abs/2505.08445 + Tue, 13 May 2025 11:13:27 +0000 - DynamicRAG: Leveraging Outputs of Large Language Model as Feedback for Dynamic Reranking in Retrieval-Augmented Generation - https://arxiv.org/abs/2505.07233 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.07233.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jiashuo Sun, Xianrui Zhong, Sizhe Zhou, Jiawei Han</p><p><b>Upvotes:</b> 2</p><p><b>Summary:</b> Retrieval-augmented generation (RAG) systems combine large language models (LLMs) with external knowledge retrieval, making them highly effective for knowledge-intensive tasks. A crucial but often under-explored component of these systems is the reranker, which refines retrieved documents to enhance generation quality and explainability. The challenge of selecting the optimal number of documents (k) remains unsolved: too few may omit critical information, while too many introduce noise and inefficiencies. Although recent studies have explored LLM-based rerankers, they primarily leverage internal model knowledge and overlook the rich supervisory signals that LLMs can provide, such as using response quality as feedback for optimizing reranking decisions. In this paper, we propose DynamicRAG, a novel RAG framework where the reranker dynamically adjusts both the order and number of retrieved documents based on the query. We model the reranker as an agent optimized through reinforcement learning (RL), using rewards derived from LLM output quality. Across seven knowledge-intensive datasets, DynamicRAG demonstrates superior performance, achieving state-of-the-art results. The model, data and code are available at https://github.com/GasolSun36/DynamicRAG</p> - https://arxiv.org/abs/2505.07233 - Mon, 12 May 2025 05:19:01 +0000 + TRAIL: Trace Reasoning and Agentic Issue Localization + https://arxiv.org/abs/2505.08638 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.08638.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Darshan Deshpande, Varun Gangal, Hersh Mehta, Jitin Krishnan, Anand Kannappan, Rebecca Qian</p><p><b>Upvotes:</b> 3</p><p><b>Summary:</b> The increasing adoption of agentic workflows across diverse domains brings a critical need to scalably and systematically evaluate the complex traces these systems generate. Current evaluation methods depend on manual, domain-specific human analysis of lengthy workflow traces - an approach that does not scale with the growing complexity and volume of agentic outputs. Error analysis in these settings is further complicated by the interplay of external tool outputs and language model reasoning, making it more challenging than traditional software debugging. In this work, we (1) articulate the need for robust and dynamic evaluation methods for agentic workflow traces, (2) introduce a formal taxonomy of error types encountered in agentic systems, and (3) present a set of 148 large human-annotated traces (TRAIL) constructed using this taxonomy and grounded in established agentic benchmarks. To ensure ecological validity, we curate traces from both single and multi-agent systems, focusing on real-world applications such as software engineering and open-world information retrieval. Our evaluations reveal that modern long context LLMs perform poorly at trace debugging, with the best Gemini-2.5-pro model scoring a mere 11% on TRAIL. Our dataset and code are made publicly available to support and accelerate future research in scalable evaluation for agentic workflows.</p> + https://arxiv.org/abs/2505.08638 + Tue, 13 May 2025 14:55:31 +0000 - UMoE: Unifying Attention and FFN with Shared Experts - https://arxiv.org/abs/2505.07260 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.07260.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yuanhang Yang, Chaozheng Wang, Jing Li</p><p><b>Upvotes:</b> 5</p><p><b>Summary:</b> Sparse Mixture of Experts (MoE) architectures have emerged as a promising approach for scaling Transformer models. While initial works primarily incorporated MoE into feed-forward network (FFN) layers, recent studies have explored extending the MoE paradigm to attention layers to enhance model performance. However, existing attention-based MoE layers require specialized implementations and demonstrate suboptimal performance compared to their FFN-based counterparts. In this paper, we aim to unify the MoE designs in attention and FFN layers by introducing a novel reformulation of the attention mechanism, revealing an underlying FFN-like structure within attention modules. Our proposed architecture, UMoE, achieves superior performance through attention-based MoE layers while enabling efficient parameter sharing between FFN and attention components.</p> - https://arxiv.org/abs/2505.07260 - Mon, 12 May 2025 06:21:44 +0000 + SkillFormer: Unified Multi-View Video Understanding for Proficiency Estimation + https://arxiv.org/abs/2505.08665 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.08665.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Edoardo Bianchi, Antonio Liotta</p><p><b>Upvotes:</b> 1</p><p><b>Summary:</b> Assessing human skill levels in complex activities is a challenging problem with applications in sports, rehabilitation, and training. In this work, we present SkillFormer, a parameter-efficient architecture for unified multi-view proficiency estimation from egocentric and exocentric videos. Building on the TimeSformer backbone, SkillFormer introduces a CrossViewFusion module that fuses view-specific features using multi-head cross-attention, learnable gating, and adaptive self-calibration. We leverage Low-Rank Adaptation to fine-tune only a small subset of parameters, significantly reducing training costs. In fact, when evaluated on the EgoExo4D dataset, SkillFormer achieves state-of-the-art accuracy in multi-view settings while demonstrating remarkable computational efficiency, using 4.5x fewer parameters and requiring 3.75x fewer training epochs than prior baselines. It excels in multiple structured tasks, confirming the value of multi-view integration for fine-grained skill assessment.</p> + https://arxiv.org/abs/2505.08665 + Tue, 13 May 2025 15:27:24 +0000 - Skywork-VL Reward: An Effective Reward Model for Multimodal Understanding and Reasoning - https://arxiv.org/abs/2505.07263 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.07263.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Xiaokun Wang, Chris, Jiangbo Pei, Wei Shen, Yi Peng, Yunzhuo Hao, Weijie Qiu, Ai Jian, Tianyidan Xie, Xuchen Song, Yang Liu, Yahui Zhou</p><p><b>Upvotes:</b> 15</p><p><b>Summary:</b> We propose Skywork-VL Reward, a multimodal reward model that provides reward signals for both multimodal understanding and reasoning tasks. Our technical approach comprises two key components: First, we construct a large-scale multimodal preference dataset that covers a wide range of tasks and scenarios, with responses collected from both standard vision-language models (VLMs) and advanced VLM reasoners. Second, we design a reward model architecture based on Qwen2.5-VL-7B-Instruct, integrating a reward head and applying multi-stage fine-tuning using pairwise ranking loss on pairwise preference data. Experimental evaluations show that Skywork-VL Reward achieves state-of-the-art results on multimodal VL-RewardBench and exhibits competitive performance on the text-only RewardBench benchmark. Furthermore, preference data constructed based on our Skywork-VL Reward proves highly effective for training Mixed Preference Optimization (MPO), leading to significant improvements in multimodal reasoning capabilities. Our results underscore Skywork-VL Reward as a significant advancement toward general-purpose, reliable reward models for multimodal alignment. Our model has been publicly released to promote transparency and reproducibility.</p> - https://arxiv.org/abs/2505.07263 - Mon, 12 May 2025 06:23:08 +0000 + NavDP: Learning Sim-to-Real Navigation Diffusion Policy with Privileged Information Guidance + https://arxiv.org/abs/2505.08712 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.08712.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Wenzhe Cai, Jiaqi Peng, Yuqiang Yang, Yujian Zhang, Meng Wei, Hanqing Wang, Yilun Chen, Tai Wang, Jiangmiao Pang</p><p><b>Upvotes:</b> 1</p><p><b>Summary:</b> Learning navigation in dynamic open-world environments is an important yet challenging skill for robots. Most previous methods rely on precise localization and mapping or learn from expensive real-world demonstrations. In this paper, we propose the Navigation Diffusion Policy (NavDP), an end-to-end framework trained solely in simulation and can zero-shot transfer to different embodiments in diverse real-world environments. The key ingredient of NavDP's network is the combination of diffusion-based trajectory generation and a critic function for trajectory selection, which are conditioned on only local observation tokens encoded from a shared policy transformer. Given the privileged information of the global environment in simulation, we scale up the demonstrations of good quality to train the diffusion policy and formulate the critic value function targets with contrastive negative samples. Our demonstration generation approach achieves about 2,500 trajectories/GPU per day, 20times more efficient than real-world data collection, and results in a large-scale navigation dataset with 363.2km trajectories across 1244 scenes. Trained with this simulation dataset, NavDP achieves state-of-the-art performance and consistently outstanding generalization capability on quadruped, wheeled, and humanoid robots in diverse indoor and outdoor environments. In addition, we present a preliminary attempt at using Gaussian Splatting to make in-domain real-to-sim fine-tuning to further bridge the sim-to-real gap. Experiments show that adding such real-to-sim data can improve the success rate by 30\% without hurting its generalization capability.</p> + https://arxiv.org/abs/2505.08712 + Tue, 13 May 2025 16:20:28 +0000 - INTELLECT-2: A Reasoning Model Trained Through Globally Decentralized Reinforcement Learning - https://arxiv.org/abs/2505.07291 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.07291.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Prime Intellect Team, Sami Jaghouar, Justus Mattern, Jack Min Ong, Jannik Straube, Manveer Basra, Aaron Pazdera, Kushal Thaman, Matthew Di Ferrante, Felix Gabriel, Fares Obeid, Kemal Erdem, Michael Keiblinger, Johannes Hagemann</p><p><b>Upvotes:</b> 2</p><p><b>Summary:</b> We introduce INTELLECT-2, the first globally distributed reinforcement learning (RL) training run of a 32 billion parameter language model. Unlike traditional centralized training efforts, INTELLECT-2 trains a reasoning model using fully asynchronous RL across a dynamic, heterogeneous swarm of permissionless compute contributors. To enable a training run with this unique infrastructure, we built various components from scratch: we introduce PRIME-RL, our training framework purpose-built for distributed asynchronous reinforcement learning, based on top of novel components such as TOPLOC, which verifies rollouts from untrusted inference workers, and SHARDCAST, which efficiently broadcasts policy weights from training nodes to inference workers. Beyond infrastructure components, we propose modifications to the standard GRPO training recipe and data filtering techniques that were crucial to achieve training stability and ensure that our model successfully learned its training objective, thus improving upon QwQ-32B, the state of the art reasoning model in the 32B parameter range. We open-source INTELLECT-2 along with all of our code and data, hoping to encourage and enable more open research in the field of decentralized training.</p> - https://arxiv.org/abs/2505.07291 - Mon, 12 May 2025 07:24:33 +0000 + Memorization-Compression Cycles Improve Generalization + https://arxiv.org/abs/2505.08727 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.08727.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Fangyuan Yu</p><p><b>Upvotes:</b> 2</p><p><b>Summary:</b> We prove theoretically that generalization improves not only through data scaling but also by compressing internal representations. To operationalize this insight, we introduce the Information Bottleneck Language Modeling (IBLM) objective, which reframes language modeling as a constrained optimization problem: minimizing representation entropy subject to optimal prediction performance. Empirically, we observe an emergent memorization-compression cycle during LLM pretraining, evidenced by oscillation positive/negative gradient alignment between cross-entropy and Matrix-Based Entropy (MBE), a measure of representation entropy. This pattern closely mirrors the predictive-compressive trade-off prescribed by IBLM and also parallels the biological alternation between awake learning and sleep consolidation. Motivated by this observation, we propose Gated Phase Transition (GAPT), a training algorithm that adaptively switches between memorization and compression phases. When applied to GPT-2 pretraining on FineWeb dataset, GAPT reduces MBE by 50% and improves cross-entropy by 4.8%. GAPT improves OOD generalizatino by 35% in a pretraining task on arithmetic multiplication. In a setting designed to simulate catastrophic forgetting, GAPT reduces interference by compressing and separating representations, achieving a 97% improvement in separation - paralleling the functional role of sleep consolidation.</p> + https://arxiv.org/abs/2505.08727 + Tue, 13 May 2025 16:37:54 +0000 - AttentionInfluence: Adopting Attention Head Influence for Weak-to-Strong Pretraining Data Selection - https://arxiv.org/abs/2505.07293 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.07293.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Kai Hua, Steven Wu, Ge Zhang, Ke Shen</p><p><b>Upvotes:</b> 17</p><p><b>Summary:</b> Recently, there has been growing interest in collecting reasoning-intensive pretraining data to improve LLMs' complex reasoning ability. Prior approaches typically rely on supervised classifiers to identify such data, which requires labeling by humans or LLMs, often introducing domain-specific biases. Due to the attention heads being crucial to in-context reasoning, we propose AttentionInfluence, a simple yet effective, training-free method without supervision signal. Our approach enables a small pretrained language model to act as a strong data selector through a simple attention head masking operation. Specifically, we identify retrieval heads and compute the loss difference when masking these heads. We apply AttentionInfluence to a 1.3B-parameter dense model to conduct data selection on the SmolLM corpus of 241B tokens, and mix the SmolLM corpus with the selected subset comprising 73B tokens to pretrain a 7B-parameter dense model using 1T training tokens and WSD learning rate scheduling. Our experimental results demonstrate substantial improvements, ranging from 1.4pp to 3.5pp, across several knowledge-intensive and reasoning-heavy benchmarks (i.e., MMLU, MMLU-Pro, AGIEval-en, GSM8K, and HumanEval). This demonstrates an effective weak-to-strong scaling property, with small models improving the final performance of larger models-offering a promising and scalable path for reasoning-centric data selection.</p> - https://arxiv.org/abs/2505.07293 - Mon, 12 May 2025 07:25:51 +0000 - - - Unified Continuous Generative Models - https://arxiv.org/abs/2505.07447 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.07447.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Peng Sun, Yi Jiang, Tao Lin</p><p><b>Upvotes:</b> 31</p><p><b>Summary:</b> Recent advances in continuous generative models, including multi-step approaches like diffusion and flow-matching (typically requiring 8-1000 sampling steps) and few-step methods such as consistency models (typically 1-8 steps), have demonstrated impressive generative performance. However, existing work often treats these approaches as distinct paradigms, resulting in separate training and sampling methodologies. We introduce a unified framework for training, sampling, and analyzing these models. Our implementation, the Unified Continuous Generative Models Trainer and Sampler (UCGM-{T,S}), achieves state-of-the-art (SOTA) performance. For example, on ImageNet 256x256 using a 675M diffusion transformer, UCGM-T trains a multi-step model achieving 1.30 FID in 20 steps and a few-step model reaching 1.42 FID in just 2 steps. Additionally, applying UCGM-S to a pre-trained model (previously 1.26 FID at 250 steps) improves performance to 1.06 FID in only 40 steps. Code is available at: https://github.com/LINs-lab/UCGM.</p> - https://arxiv.org/abs/2505.07447 - Mon, 12 May 2025 11:15:39 +0000 - - - Reinforced Internal-External Knowledge Synergistic Reasoning for Efficient Adaptive Search Agent - https://arxiv.org/abs/2505.07596 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.07596.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Ziyang Huang, Xiaowei Yuan, Yiming Ju, Jun Zhao, Kang Liu</p><p><b>Upvotes:</b> 10</p><p><b>Summary:</b> Retrieval-augmented generation (RAG) is a common strategy to reduce hallucinations in Large Language Models (LLMs). While reinforcement learning (RL) can enable LLMs to act as search agents by activating retrieval capabilities, existing ones often underutilize their internal knowledge. This can lead to redundant retrievals, potential harmful knowledge conflicts, and increased inference latency. To address these limitations, an efficient and adaptive search agent capable of discerning optimal retrieval timing and synergistically integrating parametric (internal) and retrieved (external) knowledge is in urgent need. This paper introduces the Reinforced Internal-External Knowledge Synergistic Reasoning Agent (IKEA), which could indentify its own knowledge boundary and prioritize the utilization of internal knowledge, resorting to external search only when internal knowledge is deemed insufficient. This is achieved using a novel knowledge-boundary aware reward function and a knowledge-boundary aware training dataset. These are designed for internal-external knowledge synergy oriented RL, incentivizing the model to deliver accurate answers, minimize unnecessary retrievals, and encourage appropriate external searches when its own knowledge is lacking. Evaluations across multiple knowledge reasoning tasks demonstrate that IKEA significantly outperforms baseline methods, reduces retrieval frequency significantly, and exhibits robust generalization capabilities.</p> - https://arxiv.org/abs/2505.07596 - Mon, 12 May 2025 14:21:57 +0000 - - - MiMo: Unlocking the Reasoning Potential of Language Model -- From Pretraining to Posttraining - https://arxiv.org/abs/2505.07608 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.07608.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Xiaomi LLM-Core Team, Bingquan Xia, Bowen Shen, Cici, Dawei Zhu, Di Zhang, Gang Wang, Hailin Zhang, Huaqiu Liu, Jiebao Xiao, Jinhao Dong, Liang Zhao, Peidian Li, Peng Wang, Shihua Yu, Shimao Chen, Weikun Wang, Wenhan Ma, Xiangwei Deng, Yi Huang, Yifan Song, Zihan Jiang, Bowen Ye, Can Cai, Chenhong He, Dong Zhang, Duo Zhang, Guoan Wang, Hao Tian, Haochen Zhao, Heng Qu, Hongshen Xu, Jun Shi, Kainan Bao, QingKai Fang, Kang Zhou, Kangyang Zhou, Lei Li, Menghang Zhu, Nuo Chen, Qiantong Wang, Shaohui Liu, Shicheng Li, Shuhao Gu, Shuhuai Ren, Shuo Liu, Sirui Deng, Weiji Zhuang, Weiwei Lv, Wenyu Yang, Xin Zhang, Xing Yong, Xing Zhang, Xingchen Song, Xinzhe Xu, Xu Wang, Yihan Yan, Yu Tu, Yuanyuan Tian, Yudong Wang, Yue Yu, Zhenru Lin, Zhichao Song, Zihao Yue</p><p><b>Upvotes:</b> 53</p><p><b>Summary:</b> We present MiMo-7B, a large language model born for reasoning tasks, with optimization across both pre-training and post-training stages. During pre-training, we enhance the data preprocessing pipeline and employ a three-stage data mixing strategy to strengthen the base model's reasoning potential. MiMo-7B-Base is pre-trained on 25 trillion tokens, with additional Multi-Token Prediction objective for enhanced performance and accelerated inference speed. During post-training, we curate a dataset of 130K verifiable mathematics and programming problems for reinforcement learning, integrating a test-difficulty-driven code-reward scheme to alleviate sparse-reward issues and employing strategic data resampling to stabilize training. Extensive evaluations show that MiMo-7B-Base possesses exceptional reasoning potential, outperforming even much larger 32B models. The final RL-tuned model, MiMo-7B-RL, achieves superior performance on mathematics, code and general reasoning tasks, surpassing the performance of OpenAI o1-mini. The model checkpoints are available at https://github.com/xiaomimimo/MiMo.</p> - https://arxiv.org/abs/2505.07608 - Mon, 12 May 2025 14:30:11 +0000 - - - Step1X-3D: Towards High-Fidelity and Controllable Generation of Textured 3D Assets - https://arxiv.org/abs/2505.07747 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.07747.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Weiyu Li, Xuanyang Zhang, Zheng Sun, Di Qi, Hao Li, Wei Cheng, Weiwei Cai, Shihao Wu, Jiarui Liu, Zihao Wang, Xiao Chen, Feipeng Tian, Jianxiong Pan, Zeming Li, Gang Yu, Xiangyu Zhang, Daxin Jiang, Ping Tan</p><p><b>Upvotes:</b> 46</p><p><b>Summary:</b> While generative artificial intelligence has advanced significantly across text, image, audio, and video domains, 3D generation remains comparatively underdeveloped due to fundamental challenges such as data scarcity, algorithmic limitations, and ecosystem fragmentation. To this end, we present Step1X-3D, an open framework addressing these challenges through: (1) a rigorous data curation pipeline processing >5M assets to create a 2M high-quality dataset with standardized geometric and textural properties; (2) a two-stage 3D-native architecture combining a hybrid VAE-DiT geometry generator with an diffusion-based texture synthesis module; and (3) the full open-source release of models, training code, and adaptation modules. For geometry generation, the hybrid VAE-DiT component produces TSDF representations by employing perceiver-based latent encoding with sharp edge sampling for detail preservation. The diffusion-based texture synthesis module then ensures cross-view consistency through geometric conditioning and latent-space synchronization. Benchmark results demonstrate state-of-the-art performance that exceeds existing open-source methods, while also achieving competitive quality with proprietary solutions. Notably, the framework uniquely bridges the 2D and 3D generation paradigms by supporting direct transfer of 2D control techniques~(e.g., LoRA) to 3D synthesis. By simultaneously advancing data quality, algorithmic fidelity, and reproducibility, Step1X-3D aims to establish new standards for open research in controllable 3D asset generation.</p> - https://arxiv.org/abs/2505.07747 - Mon, 12 May 2025 16:56:30 +0000 - - - Learning from Peers in Reasoning Models - https://arxiv.org/abs/2505.07787 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.07787.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Tongxu Luo, Wenyu Du, Jiaxi Bi, Stephen Chung, Zhengyang Tang, Hao Yang, Min Zhang, Benyou Wang</p><p><b>Upvotes:</b> 34</p><p><b>Summary:</b> Large Reasoning Models (LRMs) have the ability to self-correct even when they make mistakes in their reasoning paths. However, our study reveals that when the reasoning process starts with a short but poor beginning, it becomes difficult for the model to recover. We refer to this phenomenon as the "Prefix Dominance Trap". Inspired by psychological findings that peer interaction can promote self-correction without negatively impacting already accurate individuals, we propose **Learning from Peers** (LeaP) to address this phenomenon. Specifically, every tokens, each reasoning path summarizes its intermediate reasoning and shares it with others through a routing mechanism, enabling paths to incorporate peer insights during inference. However, we observe that smaller models sometimes fail to follow summarization and reflection instructions effectively. To address this, we fine-tune them into our **LeaP-T** model series. Experiments on AIME 2024, AIME 2025, AIMO 2025, and GPQA Diamond show that LeaP provides substantial improvements. For instance, QwQ-32B with LeaP achieves nearly 5 absolute points higher than the baseline on average, and surpasses DeepSeek-R1-671B on three math benchmarks with an average gain of 3.3 points. Notably, our fine-tuned LeaP-T-7B matches the performance of DeepSeek-R1-Distill-Qwen-14B on AIME 2024. In-depth analysis reveals LeaP's robust error correction by timely peer insights, showing strong error tolerance and handling varied task difficulty. LeaP marks a milestone by enabling LRMs to collaborate during reasoning. Our code, datasets, and models are available at https://learning-from-peers.github.io/ .</p> - https://arxiv.org/abs/2505.07787 - Mon, 12 May 2025 17:39:56 +0000 - - - Overflow Prevention Enhances Long-Context Recurrent LLMs - https://arxiv.org/abs/2505.07793 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.07793.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Assaf Ben-Kish, Itamar Zimerman, M. Jehanzeb Mirza, James Glass, Leonid Karlinsky, Raja Giryes</p><p><b>Upvotes:</b> 3</p><p><b>Summary:</b> A recent trend in LLMs is developing recurrent sub-quadratic models that improve long-context processing efficiency. We investigate leading large long-context models, focusing on how their fixed-size recurrent memory affects their performance. Our experiments reveal that, even when these models are trained for extended contexts, their use of long contexts remains underutilized. Specifically, we demonstrate that a chunk-based inference procedure, which identifies and processes only the most relevant portion of the input can mitigate recurrent memory failures and be effective for many long-context tasks: On LongBench, our method improves the overall performance of Falcon3-Mamba-Inst-7B by 14%, Falcon-Mamba-Inst-7B by 28%, RecurrentGemma-IT-9B by 50%, and RWKV6-Finch-7B by 51%. Surprisingly, this simple approach also leads to state-of-the-art results in the challenging LongBench v2 benchmark, showing competitive performance with equivalent size Transformers. Furthermore, our findings raise questions about whether recurrent models genuinely exploit long-range dependencies, as our single-chunk strategy delivers stronger performance - even in tasks that presumably require cross-context relations.</p> - https://arxiv.org/abs/2505.07793 - Mon, 12 May 2025 17:45:05 +0000 - - - Learning Dynamics in Continual Pre-Training for Large Language Models - https://arxiv.org/abs/2505.07796 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.07796.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Xingjin Wang, Howe Tissue, Lu Wang, Linjing Li, Daniel Dajun Zeng</p><p><b>Upvotes:</b> 12</p><p><b>Summary:</b> Continual Pre-Training (CPT) has become a popular and effective method to apply strong foundation models to specific downstream tasks. In this work, we explore the learning dynamics throughout the CPT process for large language models. We specifically focus on how general and downstream domain performance evolves at each training step, with domain performance measured via validation losses. We have observed that the CPT loss curve fundamentally characterizes the transition from one curve to another hidden curve, and could be described by decoupling the effects of distribution shift and learning rate annealing. We derive a CPT scaling law that combines the two factors, enabling the prediction of loss at any (continual) training steps and across learning rate schedules (LRS) in CPT. Our formulation presents a comprehensive understanding of several critical factors in CPT, including loss potential, peak learning rate, training steps, replay ratio, etc. Moreover, our approach can be adapted to customize training hyper-parameters to different CPT goals such as balancing general and domain-specific performance. Extensive experiments demonstrate that our scaling law holds across various CPT datasets and training hyper-parameters.</p> - https://arxiv.org/abs/2505.07796 - Mon, 12 May 2025 17:47:32 +0000 - - - Continuous Visual Autoregressive Generation via Score Maximization - https://arxiv.org/abs/2505.07812 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.07812.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Chenze Shao, Fandong Meng, Jie Zhou</p><p><b>Upvotes:</b> 3</p><p><b>Summary:</b> Conventional wisdom suggests that autoregressive models are used to process discrete data. When applied to continuous modalities such as visual data, Visual AutoRegressive modeling (VAR) typically resorts to quantization-based approaches to cast the data into a discrete space, which can introduce significant information loss. To tackle this issue, we introduce a Continuous VAR framework that enables direct visual autoregressive generation without vector quantization. The underlying theoretical foundation is strictly proper scoring rules, which provide powerful statistical tools capable of evaluating how well a generative model approximates the true distribution. Within this framework, all we need is to select a strictly proper score and set it as the training objective to optimize. We primarily explore a class of training objectives based on the energy score, which is likelihood-free and thus overcomes the difficulty of making probabilistic predictions in the continuous space. Previous efforts on continuous autoregressive generation, such as GIVT and diffusion loss, can also be derived from our framework using other strictly proper scores. Source code: https://github.com/shaochenze/EAR.</p> - https://arxiv.org/abs/2505.07812 - Mon, 12 May 2025 17:58:14 +0000 - - - DanceGRPO: Unleashing GRPO on Visual Generation - https://arxiv.org/abs/2505.07818 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.07818.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Zeyue Xue, Jie Wu, Yu Gao, Fangyuan Kong, Lingting Zhu, Mengzhao Chen, Zhiheng Liu, Wei Liu, Qiushan Guo, Weilin Huang, Ping Luo</p><p><b>Upvotes:</b> 16</p><p><b>Summary:</b> Recent breakthroughs in generative models-particularly diffusion models and rectified flows-have revolutionized visual content creation, yet aligning model outputs with human preferences remains a critical challenge. Existing reinforcement learning (RL)-based methods for visual generation face critical limitations: incompatibility with modern Ordinary Differential Equations (ODEs)-based sampling paradigms, instability in large-scale training, and lack of validation for video generation. This paper introduces DanceGRPO, the first unified framework to adapt Group Relative Policy Optimization (GRPO) to visual generation paradigms, unleashing one unified RL algorithm across two generative paradigms (diffusion models and rectified flows), three tasks (text-to-image, text-to-video, image-to-video), four foundation models (Stable Diffusion, HunyuanVideo, FLUX, SkyReel-I2V), and five reward models (image/video aesthetics, text-image alignment, video motion quality, and binary reward). To our knowledge, DanceGRPO is the first RL-based unified framework capable of seamless adaptation across diverse generative paradigms, tasks, foundational models, and reward models. DanceGRPO demonstrates consistent and substantial improvements, which outperform baselines by up to 181% on benchmarks such as HPS-v2.1, CLIP Score, VideoAlign, and GenEval. Notably, DanceGRPO not only can stabilize policy optimization for complex video generation, but also enables generative policy to better capture denoising trajectories for Best-of-N inference scaling and learn from sparse binary feedback. Our results establish DanceGRPO as a robust and versatile solution for scaling Reinforcement Learning from Human Feedback (RLHF) tasks in visual generation, offering new insights into harmonizing reinforcement learning and visual synthesis. The code will be released.</p> - https://arxiv.org/abs/2505.07818 - Mon, 12 May 2025 17:59:34 +0000 - - - H^{3}DP: Triply-Hierarchical Diffusion Policy for Visuomotor Learning - https://arxiv.org/abs/2505.07819 - <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.07819.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yiyang Lu, Yufeng Tian, Zhecheng Yuan, Xianbang Wang, Pu Hua, Zhengrong Xue, Huazhe Xu</p><p><b>Upvotes:</b> 5</p><p><b>Summary:</b> Visuomotor policy learning has witnessed substantial progress in robotic manipulation, with recent approaches predominantly relying on generative models to model the action distribution. However, these methods often overlook the critical coupling between visual perception and action prediction. In this work, we introduce Triply-Hierarchical Diffusion Policy~(H^{\mathbf{3}DP}), a novel visuomotor learning framework that explicitly incorporates hierarchical structures to strengthen the integration between visual features and action generation. H^{3}DP contains 3 levels of hierarchy: (1) depth-aware input layering that organizes RGB-D observations based on depth information; (2) multi-scale visual representations that encode semantic features at varying levels of granularity; and (3) a hierarchically conditioned diffusion process that aligns the generation of coarse-to-fine actions with corresponding visual features. Extensive experiments demonstrate that H^{3}DP yields a +27.5% average relative improvement over baselines across 44 simulation tasks and achieves superior performance in 4 challenging bimanual real-world manipulation tasks. Project Page: https://lyy-iiis.github.io/h3dp/.</p> - https://arxiv.org/abs/2505.07819 - Mon, 12 May 2025 17:59:43 +0000 + Aya Vision: Advancing the Frontier of Multilingual Multimodality + https://arxiv.org/abs/2505.08751 + <p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2505.08751.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Saurabh Dash, Yiyang Nan, John Dang, Arash Ahmadian, Shivalika Singh, Madeline Smith, Bharat Venkitesh, Vlad Shmyhlo, Viraat Aryabumi, Walter Beller-Morales, Jeremy Pekmez, Jason Ozuzu, Pierre Richemond, Acyr Locatelli, Nick Frosst, Phil Blunsom, Aidan Gomez, Ivan Zhang, Marzieh Fadaee, Manoj Govindassamy, Sudip Roy, Matthias Gallé, Beyza Ermis, Ahmet Üstün, Sara Hooker</p><p><b>Upvotes:</b> 2</p><p><b>Summary:</b> Building multimodal language models is fundamentally challenging: it requires aligning vision and language modalities, curating high-quality instruction data, and avoiding the degradation of existing text-only capabilities once vision is introduced. These difficulties are further magnified in the multilingual setting, where the need for multimodal data in different languages exacerbates existing data scarcity, machine translation often distorts meaning, and catastrophic forgetting is more pronounced. To address the aforementioned challenges, we introduce novel techniques spanning both data and modeling. First, we develop a synthetic annotation framework that curates high-quality, diverse multilingual multimodal instruction data, enabling Aya Vision models to produce natural, human-preferred responses to multimodal inputs across many languages. Complementing this, we propose a cross-modal model merging technique that mitigates catastrophic forgetting, effectively preserving text-only capabilities while simultaneously enhancing multimodal generative performance. Aya-Vision-8B achieves best-in-class performance compared to strong multimodal models such as Qwen-2.5-VL-7B, Pixtral-12B, and even much larger Llama-3.2-90B-Vision. We further scale this approach with Aya-Vision-32B, which outperforms models more than twice its size, such as Molmo-72B and LLaMA-3.2-90B-Vision. Our work advances multilingual progress on the multi-modal frontier, and provides insights into techniques that effectively bend the need for compute while delivering extremely high performance.</p> + https://arxiv.org/abs/2505.08751 + Tue, 13 May 2025 17:03:48 +0000