bot: update RSS feed
This commit is contained in:
github-actions[bot]
42
feed.xml
42
feed.xml
@@ -7,32 +7,32 @@
|
|||||||
<docs>http://www.rssboard.org/rss-specification</docs>
|
<docs>http://www.rssboard.org/rss-specification</docs>
|
||||||
<generator>python-feedgen</generator>
|
<generator>python-feedgen</generator>
|
||||||
<language>en</language>
|
<language>en</language>
|
||||||
<lastBuildDate>Sat, 28 Mar 2026 00:05:38 +0000</lastBuildDate>
|
<lastBuildDate>Sun, 29 Mar 2026 00:06:11 +0000</lastBuildDate>
|
||||||
<item>
|
<item>
|
||||||
<title>MSA: Memory Sparse Attention for Efficient End-to-End Memory Model Scaling to 100M Tokens</title>
|
<title>MSA: Memory Sparse Attention for Efficient End-to-End Memory Model Scaling to 100M Tokens</title>
|
||||||
<link>https://arxiv.org/abs/2603.23516</link>
|
<link>https://arxiv.org/abs/2603.23516</link>
|
||||||
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.23516.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yu Chen, Runkai Chen, Sheng Yi, Xinda Zhao, Xiaohong Li, Jianjin Zhang, Jun Sun, Chuanrui Hu, Yunyun Han, Lidong Bing, Yafeng Deng, Tianqiao Chen</p><p><b>Upvotes:</b> 19</p><p><b>Summary:</b> Long-term memory is a cornerstone of human intelligence. Enabling AI to process lifetime-scale information remains a long-standing pursuit in the field. Due to the constraints of full-attention architectures, the effective context length of large language models (LLMs) is typically limited to 1M tokens. Existing approaches, such as hybrid linear attention, fixed-size memory states (e.g., RNNs), and external storage methods like RAG or agent systems, attempt to extend this limit. However, they often suffer from severe precision degradation and rapidly increasing latency as context length grows, an inability to dynamically modify memory content, or a lack of end-to-end optimization. These bottlenecks impede complex scenarios like large-corpus summarization, Digital Twins, and long-history agent reasoning, while limiting memory capacity and slowing inference. We present Memory Sparse Attention (MSA), an end-to-end trainable, efficient, and massively scalable memory model framework. Through core innovations including scalable sparse attention and document-wise RoPE, MSA achieves linear complexity in both training and inference while maintaining exceptional stability, exhibiting less than 9% degradation when scaling from 16K to 100M tokens. Furthermore, KV cache compression, combined with Memory Parallel, enables 100M-token inference on 2xA800 GPUs. We also propose Memory Interleaving to facilitate complex multi-hop reasoning across scattered memory segments. MSA significantly surpasses frontier LLMs, state-of-the-art RAG systems, and leading memory agents in long-context benchmarks. These results demonstrate that by decoupling memory capacity from reasoning, MSA provides a scalable foundation to endow general-purpose models with intrinsic, lifetime-scale memory.</p></description>
|
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.23516.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yu Chen, Runkai Chen, Sheng Yi, Xinda Zhao, Xiaohong Li, Jianjin Zhang, Jun Sun, Chuanrui Hu, Yunyun Han, Lidong Bing, Yafeng Deng, Tianqiao Chen</p><p><b>Upvotes:</b> 27</p><p><b>Summary:</b> Long-term memory is a cornerstone of human intelligence. Enabling AI to process lifetime-scale information remains a long-standing pursuit in the field. Due to the constraints of full-attention architectures, the effective context length of large language models (LLMs) is typically limited to 1M tokens. Existing approaches, such as hybrid linear attention, fixed-size memory states (e.g., RNNs), and external storage methods like RAG or agent systems, attempt to extend this limit. However, they often suffer from severe precision degradation and rapidly increasing latency as context length grows, an inability to dynamically modify memory content, or a lack of end-to-end optimization. These bottlenecks impede complex scenarios like large-corpus summarization, Digital Twins, and long-history agent reasoning, while limiting memory capacity and slowing inference. We present Memory Sparse Attention (MSA), an end-to-end trainable, efficient, and massively scalable memory model framework. Through core innovations including scalable sparse attention and document-wise RoPE, MSA achieves linear complexity in both training and inference while maintaining exceptional stability, exhibiting less than 9% degradation when scaling from 16K to 100M tokens. Furthermore, KV cache compression, combined with Memory Parallel, enables 100M-token inference on 2xA800 GPUs. We also propose Memory Interleaving to facilitate complex multi-hop reasoning across scattered memory segments. MSA significantly surpasses frontier LLMs, state-of-the-art RAG systems, and leading memory agents in long-context benchmarks. These results demonstrate that by decoupling memory capacity from reasoning, MSA provides a scalable foundation to endow general-purpose models with intrinsic, lifetime-scale memory.</p></description>
|
||||||
<guid isPermaLink="false">https://arxiv.org/abs/2603.23516</guid>
|
<guid isPermaLink="false">https://arxiv.org/abs/2603.23516</guid>
|
||||||
<pubDate>Fri, 06 Mar 2026 02:29:54 +0000</pubDate>
|
<pubDate>Fri, 06 Mar 2026 02:29:54 +0000</pubDate>
|
||||||
</item>
|
</item>
|
||||||
<item>
|
<item>
|
||||||
<title>Pixel-level Scene Understanding in One Token: Visual States Need What-is-Where Composition</title>
|
<title>Pixel-level Scene Understanding in One Token: Visual States Need What-is-Where Composition</title>
|
||||||
<link>https://arxiv.org/abs/2603.13904</link>
|
<link>https://arxiv.org/abs/2603.13904</link>
|
||||||
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.13904.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Seokmin Lee, Yunghee Lee, Byeonghyun Pak, Byeongju Woo</p><p><b>Upvotes:</b> 1</p><p><b>Summary:</b> For robotic agents operating in dynamic environments, learning visual state representations from streaming video observations is essential for sequential decision making. Recent self-supervised learning methods have shown strong transferability across vision tasks, but they do not explicitly address what a good visual state should encode. We argue that effective visual states must capture what-is-where by jointly encoding the semantic identities of scene elements and their spatial locations, enabling reliable detection of subtle dynamics across observations. To this end, we propose CroBo, a visual state representation learning framework based on a global-to-local reconstruction objective. Given a reference observation compressed into a compact bottleneck token, CroBo learns to reconstruct heavily masked patches in a local target crop from sparse visible cues, using the global bottleneck token as context. This learning objective encourages the bottleneck token to encode a fine-grained representation of scene-wide semantic entities, including their identities, spatial locations, and configurations. As a result, the learned visual states reveal how scene elements move and interact over time, supporting sequential decision making. We evaluate CroBo on diverse vision-based robot policy learning benchmarks, where it achieves state-of-the-art performance. Reconstruction analyses and perceptual straightness experiments further show that the learned representations preserve pixel-level scene composition and encode what-moves-where across observations. Project page available at: https://seokminlee-chris.github.io/CroBo-ProjectPage.</p></description>
|
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.13904.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Seokmin Lee, Yunghee Lee, Byeonghyun Pak, Byeongju Woo</p><p><b>Upvotes:</b> 2</p><p><b>Summary:</b> For robotic agents operating in dynamic environments, learning visual state representations from streaming video observations is essential for sequential decision making. Recent self-supervised learning methods have shown strong transferability across vision tasks, but they do not explicitly address what a good visual state should encode. We argue that effective visual states must capture what-is-where by jointly encoding the semantic identities of scene elements and their spatial locations, enabling reliable detection of subtle dynamics across observations. To this end, we propose CroBo, a visual state representation learning framework based on a global-to-local reconstruction objective. Given a reference observation compressed into a compact bottleneck token, CroBo learns to reconstruct heavily masked patches in a local target crop from sparse visible cues, using the global bottleneck token as context. This learning objective encourages the bottleneck token to encode a fine-grained representation of scene-wide semantic entities, including their identities, spatial locations, and configurations. As a result, the learned visual states reveal how scene elements move and interact over time, supporting sequential decision making. We evaluate CroBo on diverse vision-based robot policy learning benchmarks, where it achieves state-of-the-art performance. Reconstruction analyses and perceptual straightness experiments further show that the learned representations preserve pixel-level scene composition and encode what-moves-where across observations. Project page available at: https://seokminlee-chris.github.io/CroBo-ProjectPage.</p></description>
|
||||||
<guid isPermaLink="false">https://arxiv.org/abs/2603.13904</guid>
|
<guid isPermaLink="false">https://arxiv.org/abs/2603.13904</guid>
|
||||||
<pubDate>Sat, 14 Mar 2026 11:32:02 +0000</pubDate>
|
<pubDate>Sat, 14 Mar 2026 11:32:02 +0000</pubDate>
|
||||||
</item>
|
</item>
|
||||||
<item>
|
<item>
|
||||||
<title>Representation Alignment for Just Image Transformers is not Easier than You Think</title>
|
<title>Representation Alignment for Just Image Transformers is not Easier than You Think</title>
|
||||||
<link>https://arxiv.org/abs/2603.14366</link>
|
<link>https://arxiv.org/abs/2603.14366</link>
|
||||||
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.14366.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jaeyo Shin, Jiwook Kim, Hyunjung Shim</p><p><b>Upvotes:</b> 5</p><p><b>Summary:</b> Representation Alignment (REPA) has emerged as a simple way to accelerate Diffusion Transformers training in latent space. At the same time, pixel-space diffusion transformers such as Just image Transformers (JiT) have attracted growing attention because they remove a dependency on a pretrained tokenizer, and then avoid the reconstruction bottleneck of latent diffusion. This paper shows that the REPA can fail for JiT. REPA yields worse FID for JiT as training proceeds and collapses diversity on image subsets that are tightly clustered in the representation space of pretrained semantic encoder on ImageNet. We trace the failure to an information asymmetry: denoising occurs in the high dimensional image space, while the semantic target is strongly compressed, making direct regression a shortcut objective. We propose PixelREPA, which transforms the alignment target and constrains alignment with a Masked Transformer Adapter that combines a shallow transformer adapter with partial token masking. PixelREPA improves both training convergence and final quality. PixelREPA reduces FID from 3.66 to 3.17 for JiT-B/16 and improves Inception Score (IS) from 275.1 to 284.6 on ImageNet 256 times 256, while achieving > 2times faster convergence. Finally, PixelREPA-H/16 achieves FID=1.81 and IS=317.2. Our code is available at https://github.com/kaist-cvml/PixelREPA.</p></description>
|
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.14366.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jaeyo Shin, Jiwook Kim, Hyunjung Shim</p><p><b>Upvotes:</b> 6</p><p><b>Summary:</b> Representation Alignment (REPA) has emerged as a simple way to accelerate Diffusion Transformers training in latent space. At the same time, pixel-space diffusion transformers such as Just image Transformers (JiT) have attracted growing attention because they remove a dependency on a pretrained tokenizer, and then avoid the reconstruction bottleneck of latent diffusion. This paper shows that the REPA can fail for JiT. REPA yields worse FID for JiT as training proceeds and collapses diversity on image subsets that are tightly clustered in the representation space of pretrained semantic encoder on ImageNet. We trace the failure to an information asymmetry: denoising occurs in the high dimensional image space, while the semantic target is strongly compressed, making direct regression a shortcut objective. We propose PixelREPA, which transforms the alignment target and constrains alignment with a Masked Transformer Adapter that combines a shallow transformer adapter with partial token masking. PixelREPA improves both training convergence and final quality. PixelREPA reduces FID from 3.66 to 3.17 for JiT-B/16 and improves Inception Score (IS) from 275.1 to 284.6 on ImageNet 256 times 256, while achieving > 2times faster convergence. Finally, PixelREPA-H/16 achieves FID=1.81 and IS=317.2. Our code is available at https://github.com/kaist-cvml/PixelREPA.</p></description>
|
||||||
<guid isPermaLink="false">https://arxiv.org/abs/2603.14366</guid>
|
<guid isPermaLink="false">https://arxiv.org/abs/2603.14366</guid>
|
||||||
<pubDate>Sun, 15 Mar 2026 13:08:31 +0000</pubDate>
|
<pubDate>Sun, 15 Mar 2026 13:08:31 +0000</pubDate>
|
||||||
</item>
|
</item>
|
||||||
<item>
|
<item>
|
||||||
<title>Nudging Hidden States: Training-Free Model Steering for Chain-of-Thought Reasoning in Large Audio-Language Models</title>
|
<title>Nudging Hidden States: Training-Free Model Steering for Chain-of-Thought Reasoning in Large Audio-Language Models</title>
|
||||||
<link>https://arxiv.org/abs/2603.14636</link>
|
<link>https://arxiv.org/abs/2603.14636</link>
|
||||||
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.14636.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Lok-Lam Ieong, Chia-Chien Chen, Chih-Kai Yang, Yu-Han Huang, An-Yu Cheng, Hung-yi Lee</p><p><b>Upvotes:</b> 0</p><p><b>Summary:</b> Chain-of-thought (CoT) prompting has been extended to large audio-language models (LALMs) to elicit reasoning, yet enhancing its effectiveness without training remains challenging. We study inference-time model steering as a training-free approach to improve LALM reasoning. We introduce three strategies using diverse information sources and evaluate them across four LALMs and four benchmarks. Results show general accuracy gains up to 4.4% over CoT prompting. Notably, we identify a cross-modal transfer where steering vectors derived from few text samples effectively guide speech-based reasoning, demonstrating high data efficiency. We also examine hyperparameter sensitivity to understand the robustness of these approaches. Our findings position model steering as a practical direction for strengthening LALM reasoning.</p></description>
|
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.14636.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Lok-Lam Ieong, Chia-Chien Chen, Chih-Kai Yang, Yu-Han Huang, An-Yu Cheng, Hung-yi Lee</p><p><b>Upvotes:</b> 1</p><p><b>Summary:</b> Chain-of-thought (CoT) prompting has been extended to large audio-language models (LALMs) to elicit reasoning, yet enhancing its effectiveness without training remains challenging. We study inference-time model steering as a training-free approach to improve LALM reasoning. We introduce three strategies using diverse information sources and evaluate them across four LALMs and four benchmarks. Results show general accuracy gains up to 4.4% over CoT prompting. Notably, we identify a cross-modal transfer where steering vectors derived from few text samples effectively guide speech-based reasoning, demonstrating high data efficiency. We also examine hyperparameter sensitivity to understand the robustness of these approaches. Our findings position model steering as a practical direction for strengthening LALM reasoning.</p></description>
|
||||||
<guid isPermaLink="false">https://arxiv.org/abs/2603.14636</guid>
|
<guid isPermaLink="false">https://arxiv.org/abs/2603.14636</guid>
|
||||||
<pubDate>Sun, 15 Mar 2026 22:12:36 +0000</pubDate>
|
<pubDate>Sun, 15 Mar 2026 22:12:36 +0000</pubDate>
|
||||||
</item>
|
</item>
|
||||||
@@ -46,7 +46,7 @@
|
|||||||
<item>
|
<item>
|
||||||
<title>MemMA: Coordinating the Memory Cycle through Multi-Agent Reasoning and In-Situ Self-Evolution</title>
|
<title>MemMA: Coordinating the Memory Cycle through Multi-Agent Reasoning and In-Situ Self-Evolution</title>
|
||||||
<link>https://arxiv.org/abs/2603.18718</link>
|
<link>https://arxiv.org/abs/2603.18718</link>
|
||||||
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.18718.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Minhua Lin, Zhiwei Zhang, Hanqing Lu, Hui Liu, Xianfeng Tang, Qi He, Xiang Zhang, Suhang Wang</p><p><b>Upvotes:</b> 2</p><p><b>Summary:</b> Memory-augmented LLM agents maintain external memory banks to support long-horizon interaction, yet most existing systems treat construction, retrieval, and utilization as isolated subroutines. This creates two coupled challenges: strategic blindness on the forward path of the memory cycle, where construction and retrieval are driven by local heuristics rather than explicit strategic reasoning, and sparse, delayed supervision on the backward path, where downstream failures rarely translate into direct repairs of the memory bank. To address these challenges, we propose MemMA, a plug-and-play multi-agent framework that coordinates the memory cycle along both the forward and backward paths. On the forward path, a Meta-Thinker produces structured guidance that steers a Memory Manager during construction and directs a Query Reasoner during iterative retrieval. On the backward path, MemMA introduces in-situ self-evolving memory construction, which synthesizes probe QA pairs, verifies the current memory, and converts failures into repair actions before the memory is finalized. Extensive experiments on LoCoMo show that MemMA consistently outperforms existing baselines across multiple LLM backbones and improves three different storage backends in a plug-and-play manner. Our code is publicly available at https://github.com/ventr1c/memma.</p></description>
|
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.18718.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Minhua Lin, Zhiwei Zhang, Hanqing Lu, Hui Liu, Xianfeng Tang, Qi He, Xiang Zhang, Suhang Wang</p><p><b>Upvotes:</b> 6</p><p><b>Summary:</b> Memory-augmented LLM agents maintain external memory banks to support long-horizon interaction, yet most existing systems treat construction, retrieval, and utilization as isolated subroutines. This creates two coupled challenges: strategic blindness on the forward path of the memory cycle, where construction and retrieval are driven by local heuristics rather than explicit strategic reasoning, and sparse, delayed supervision on the backward path, where downstream failures rarely translate into direct repairs of the memory bank. To address these challenges, we propose MemMA, a plug-and-play multi-agent framework that coordinates the memory cycle along both the forward and backward paths. On the forward path, a Meta-Thinker produces structured guidance that steers a Memory Manager during construction and directs a Query Reasoner during iterative retrieval. On the backward path, MemMA introduces in-situ self-evolving memory construction, which synthesizes probe QA pairs, verifies the current memory, and converts failures into repair actions before the memory is finalized. Extensive experiments on LoCoMo show that MemMA consistently outperforms existing baselines across multiple LLM backbones and improves three different storage backends in a plug-and-play manner. Our code is publicly available at https://github.com/ventr1c/memma.</p></description>
|
||||||
<guid isPermaLink="false">https://arxiv.org/abs/2603.18718</guid>
|
<guid isPermaLink="false">https://arxiv.org/abs/2603.18718</guid>
|
||||||
<pubDate>Thu, 19 Mar 2026 10:15:59 +0000</pubDate>
|
<pubDate>Thu, 19 Mar 2026 10:15:59 +0000</pubDate>
|
||||||
</item>
|
</item>
|
||||||
@@ -67,42 +67,42 @@
|
|||||||
<item>
|
<item>
|
||||||
<title>AVO: Agentic Variation Operators for Autonomous Evolutionary Search</title>
|
<title>AVO: Agentic Variation Operators for Autonomous Evolutionary Search</title>
|
||||||
<link>https://arxiv.org/abs/2603.24517</link>
|
<link>https://arxiv.org/abs/2603.24517</link>
|
||||||
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.24517.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Terry Chen, Zhifan Ye, Bing Xu, Zihao Ye, Timmy Liu, Ali Hassani, Tianqi Chen, Andrew Kerr, Haicheng Wu, Yang Xu, Yu-Jung Chen, Hanfeng Chen, Aditya Kane, Ronny Krashinsky, Ming-Yu Liu, Vinod Grover, Luis Ceze, Roger Bringmann, John Tran, Wei Liu, Fung Xie, Michael Lightstone, Humphrey Shi</p><p><b>Upvotes:</b> 3</p><p><b>Summary:</b> Agentic Variation Operators (AVO) are a new family of evolutionary variation operators that replace the fixed mutation, crossover, and hand-designed heuristics of classical evolutionary search with autonomous coding agents. Rather than confining a language model to candidate generation within a prescribed pipeline, AVO instantiates variation as a self-directed agent loop that can consult the current lineage, a domain-specific knowledge base, and execution feedback to propose, repair, critique, and verify implementation edits. We evaluate AVO on attention, among the most aggressively optimized kernel targets in AI, on NVIDIA Blackwell (B200) GPUs. Over 7 days of continuous autonomous evolution on multi-head attention, AVO discovers kernels that outperform cuDNN by up to 3.5% and FlashAttention-4 by up to 10.5% across the evaluated configurations. The discovered optimizations transfer readily to grouped-query attention, requiring only 30 minutes of additional autonomous adaptation and yielding gains of up to 7.0% over cuDNN and 9.3% over FlashAttention-4. Together, these results show that agentic variation operators move beyond prior LLM-in-the-loop evolutionary pipelines by elevating the agent from candidate generator to variation operator, and can discover performance-critical micro-architectural optimizations that produce kernels surpassing state-of-the-art expert-engineered attention implementations on today's most advanced GPU hardware.</p></description>
|
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.24517.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Terry Chen, Zhifan Ye, Bing Xu, Zihao Ye, Timmy Liu, Ali Hassani, Tianqi Chen, Andrew Kerr, Haicheng Wu, Yang Xu, Yu-Jung Chen, Hanfeng Chen, Aditya Kane, Ronny Krashinsky, Ming-Yu Liu, Vinod Grover, Luis Ceze, Roger Bringmann, John Tran, Wei Liu, Fung Xie, Michael Lightstone, Humphrey Shi</p><p><b>Upvotes:</b> 5</p><p><b>Summary:</b> Agentic Variation Operators (AVO) are a new family of evolutionary variation operators that replace the fixed mutation, crossover, and hand-designed heuristics of classical evolutionary search with autonomous coding agents. Rather than confining a language model to candidate generation within a prescribed pipeline, AVO instantiates variation as a self-directed agent loop that can consult the current lineage, a domain-specific knowledge base, and execution feedback to propose, repair, critique, and verify implementation edits. We evaluate AVO on attention, among the most aggressively optimized kernel targets in AI, on NVIDIA Blackwell (B200) GPUs. Over 7 days of continuous autonomous evolution on multi-head attention, AVO discovers kernels that outperform cuDNN by up to 3.5% and FlashAttention-4 by up to 10.5% across the evaluated configurations. The discovered optimizations transfer readily to grouped-query attention, requiring only 30 minutes of additional autonomous adaptation and yielding gains of up to 7.0% over cuDNN and 9.3% over FlashAttention-4. Together, these results show that agentic variation operators move beyond prior LLM-in-the-loop evolutionary pipelines by elevating the agent from candidate generator to variation operator, and can discover performance-critical micro-architectural optimizations that produce kernels surpassing state-of-the-art expert-engineered attention implementations on today's most advanced GPU hardware.</p></description>
|
||||||
<guid isPermaLink="false">https://arxiv.org/abs/2603.24517</guid>
|
<guid isPermaLink="false">https://arxiv.org/abs/2603.24517</guid>
|
||||||
<pubDate>Wed, 25 Mar 2026 16:55:04 +0000</pubDate>
|
<pubDate>Wed, 25 Mar 2026 16:55:04 +0000</pubDate>
|
||||||
</item>
|
</item>
|
||||||
<item>
|
<item>
|
||||||
<title>VFIG: Vectorizing Complex Figures in SVG with Vision-Language Models</title>
|
<title>VFIG: Vectorizing Complex Figures in SVG with Vision-Language Models</title>
|
||||||
<link>https://arxiv.org/abs/2603.24575</link>
|
<link>https://arxiv.org/abs/2603.24575</link>
|
||||||
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.24575.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Qijia He, Xunmei Liu, Hammaad Memon, Ziang Li, Zixian Ma, Jaemin Cho, Jason Ren, Daniel S Weld, Ranjay Krishna</p><p><b>Upvotes:</b> 0</p><p><b>Summary:</b> Scalable Vector Graphics (SVG) are an essential format for technical illustration and digital design, offering precise resolution independence and flexible semantic editability. In practice, however, original vector source files are frequently lost or inaccessible, leaving only "flat" rasterized versions (e.g., PNG or JPEG) that are difficult to modify or scale. Manually reconstructing these figures is a prohibitively labor-intensive process, requiring specialized expertise to recover the original geometric intent. To bridge this gap, we propose VFIG, a family of Vision-Language Models trained for complex and high-fidelity figure-to-SVG conversion. While this task is inherently data-driven, existing datasets are typically small-scale and lack the complexity of professional diagrams. We address this by introducing VFIG-DATA, a large-scale dataset of 66K high-quality figure-SVG pairs, curated from a diverse mix of real-world paper figures and procedurally generated diagrams. Recognizing that SVGs are composed of recurring primitives and hierarchical local structures, we introduce a coarse-to-fine training curriculum that begins with supervised fine-tuning (SFT) to learn atomic primitives and transitions to reinforcement learning (RL) refinement to optimize global diagram fidelity, layout consistency, and topological edge cases. Finally, we introduce VFIG-BENCH, a comprehensive evaluation suite with novel metrics designed to measure the structural integrity of complex figures. VFIG achieves state-of-the-art performance among open-source models and performs on par with GPT-5.2, achieving a VLM-Judge score of 0.829 on VFIG-BENCH.</p></description>
|
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.24575.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Qijia He, Xunmei Liu, Hammaad Memon, Ziang Li, Zixian Ma, Jaemin Cho, Jason Ren, Daniel S Weld, Ranjay Krishna</p><p><b>Upvotes:</b> 11</p><p><b>Summary:</b> Scalable Vector Graphics (SVG) are an essential format for technical illustration and digital design, offering precise resolution independence and flexible semantic editability. In practice, however, original vector source files are frequently lost or inaccessible, leaving only "flat" rasterized versions (e.g., PNG or JPEG) that are difficult to modify or scale. Manually reconstructing these figures is a prohibitively labor-intensive process, requiring specialized expertise to recover the original geometric intent. To bridge this gap, we propose VFIG, a family of Vision-Language Models trained for complex and high-fidelity figure-to-SVG conversion. While this task is inherently data-driven, existing datasets are typically small-scale and lack the complexity of professional diagrams. We address this by introducing VFIG-DATA, a large-scale dataset of 66K high-quality figure-SVG pairs, curated from a diverse mix of real-world paper figures and procedurally generated diagrams. Recognizing that SVGs are composed of recurring primitives and hierarchical local structures, we introduce a coarse-to-fine training curriculum that begins with supervised fine-tuning (SFT) to learn atomic primitives and transitions to reinforcement learning (RL) refinement to optimize global diagram fidelity, layout consistency, and topological edge cases. Finally, we introduce VFIG-BENCH, a comprehensive evaluation suite with novel metrics designed to measure the structural integrity of complex figures. VFIG achieves state-of-the-art performance among open-source models and performs on par with GPT-5.2, achieving a VLM-Judge score of 0.829 on VFIG-BENCH.</p></description>
|
||||||
<guid isPermaLink="false">https://arxiv.org/abs/2603.24575</guid>
|
<guid isPermaLink="false">https://arxiv.org/abs/2603.24575</guid>
|
||||||
<pubDate>Wed, 25 Mar 2026 17:52:23 +0000</pubDate>
|
<pubDate>Wed, 25 Mar 2026 17:52:23 +0000</pubDate>
|
||||||
</item>
|
</item>
|
||||||
<item>
|
<item>
|
||||||
<title>SlopCodeBench: Benchmarking How Coding Agents Degrade Over Long-Horizon Iterative Tasks</title>
|
<title>SlopCodeBench: Benchmarking How Coding Agents Degrade Over Long-Horizon Iterative Tasks</title>
|
||||||
<link>https://arxiv.org/abs/2603.24755</link>
|
<link>https://arxiv.org/abs/2603.24755</link>
|
||||||
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.24755.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Gabriel Orlanski, Devjeet Roy, Alexander Yun, Changho Shin, Alex Gu, Albert Ge, Dyah Adila, Frederic Sala, Aws Albarghouthi</p><p><b>Upvotes:</b> 19</p><p><b>Summary:</b> Software development is iterative, yet agentic coding benchmarks overwhelmingly evaluate single-shot solutions against complete specifications. Code can pass the test suite but become progressively harder to extend. Recent iterative benchmarks attempt to close this gap, but constrain the agent's design decisions too tightly to faithfully measure how code quality shapes future extensions. We introduce SlopCodeBench, a language-agnostic benchmark comprising 20 problems and 93 checkpoints, in which agents repeatedly extend their own prior solutions under evolving specifications that force architectural decisions without prescribing internal structure. We track two trajectory-level quality signals: verbosity, the fraction of redundant or duplicated code, and structural erosion, the share of complexity mass concentrated in high-complexity functions. No agent solves any problem end-to-end across 11 models; the highest checkpoint solve rate is 17.2%. Quality degrades steadily: erosion rises in 80% of trajectories and verbosity in 89.8%. Against 48 open-source Python repositories, agent code is 2.2x more verbose and markedly more eroded. Tracking 20 of those repositories over time shows that human code stays flat, while agent code deteriorates with each iteration. A prompt-intervention study shows that initial quality can be improved, but it does not halt degradation. These results demonstrate that pass-rate benchmarks systematically undermeasure extension robustness, and that current agents lack the design discipline iterative software development demands.</p></description>
|
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.24755.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Gabriel Orlanski, Devjeet Roy, Alexander Yun, Changho Shin, Alex Gu, Albert Ge, Dyah Adila, Frederic Sala, Aws Albarghouthi</p><p><b>Upvotes:</b> 22</p><p><b>Summary:</b> Software development is iterative, yet agentic coding benchmarks overwhelmingly evaluate single-shot solutions against complete specifications. Code can pass the test suite but become progressively harder to extend. Recent iterative benchmarks attempt to close this gap, but constrain the agent's design decisions too tightly to faithfully measure how code quality shapes future extensions. We introduce SlopCodeBench, a language-agnostic benchmark comprising 20 problems and 93 checkpoints, in which agents repeatedly extend their own prior solutions under evolving specifications that force architectural decisions without prescribing internal structure. We track two trajectory-level quality signals: verbosity, the fraction of redundant or duplicated code, and structural erosion, the share of complexity mass concentrated in high-complexity functions. No agent solves any problem end-to-end across 11 models; the highest checkpoint solve rate is 17.2%. Quality degrades steadily: erosion rises in 80% of trajectories and verbosity in 89.8%. Against 48 open-source Python repositories, agent code is 2.2x more verbose and markedly more eroded. Tracking 20 of those repositories over time shows that human code stays flat, while agent code deteriorates with each iteration. A prompt-intervention study shows that initial quality can be improved, but it does not halt degradation. These results demonstrate that pass-rate benchmarks systematically undermeasure extension robustness, and that current agents lack the design discipline iterative software development demands.</p></description>
|
||||||
<guid isPermaLink="false">https://arxiv.org/abs/2603.24755</guid>
|
<guid isPermaLink="false">https://arxiv.org/abs/2603.24755</guid>
|
||||||
<pubDate>Wed, 25 Mar 2026 19:26:44 +0000</pubDate>
|
<pubDate>Wed, 25 Mar 2026 19:26:44 +0000</pubDate>
|
||||||
</item>
|
</item>
|
||||||
<item>
|
<item>
|
||||||
<title>AVControl: Efficient Framework for Training Audio-Visual Controls</title>
|
<title>AVControl: Efficient Framework for Training Audio-Visual Controls</title>
|
||||||
<link>https://arxiv.org/abs/2603.24793</link>
|
<link>https://arxiv.org/abs/2603.24793</link>
|
||||||
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.24793.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Matan Ben-Yosef, Tavi Halperin, Naomi Ken Korem, Mohammad Salama, Harel Cain, Asaf Joseph, Anthony Chen, Urska Jelercic, Ofir Bibi</p><p><b>Upvotes:</b> 10</p><p><b>Summary:</b> Controlling video and audio generation requires diverse modalities, from depth and pose to camera trajectories and audio transformations, yet existing approaches either train a single monolithic model for a fixed set of controls or introduce costly architectural changes for each new modality. We introduce AVControl, a lightweight, extendable framework built on LTX-2, a joint audio-visual foundation model, where each control modality is trained as a separate LoRA on a parallel canvas that provides the reference signal as additional tokens in the attention layers, requiring no architectural changes beyond the LoRA adapters themselves. We show that simply extending image-based in-context methods to video fails for structural control, and that our parallel canvas approach resolves this. On the VACE Benchmark, we outperform all evaluated baselines on depth- and pose-guided generation, inpainting, and outpainting, and show competitive results on camera control and audio-visual benchmarks. Our framework supports a diverse set of independently trained modalities: spatially-aligned controls such as depth, pose, and edges, camera trajectory with intrinsics, sparse motion control, video editing, and, to our knowledge, the first modular audio-visual controls for a joint generation model. Our method is both compute- and data-efficient: each modality requires only a small dataset and converges within a few hundred to a few thousand training steps, a fraction of the budget of monolithic alternatives. We publicly release our code and trained LoRA checkpoints.</p></description>
|
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.24793.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Matan Ben-Yosef, Tavi Halperin, Naomi Ken Korem, Mohammad Salama, Harel Cain, Asaf Joseph, Anthony Chen, Urska Jelercic, Ofir Bibi</p><p><b>Upvotes:</b> 15</p><p><b>Summary:</b> Controlling video and audio generation requires diverse modalities, from depth and pose to camera trajectories and audio transformations, yet existing approaches either train a single monolithic model for a fixed set of controls or introduce costly architectural changes for each new modality. We introduce AVControl, a lightweight, extendable framework built on LTX-2, a joint audio-visual foundation model, where each control modality is trained as a separate LoRA on a parallel canvas that provides the reference signal as additional tokens in the attention layers, requiring no architectural changes beyond the LoRA adapters themselves. We show that simply extending image-based in-context methods to video fails for structural control, and that our parallel canvas approach resolves this. On the VACE Benchmark, we outperform all evaluated baselines on depth- and pose-guided generation, inpainting, and outpainting, and show competitive results on camera control and audio-visual benchmarks. Our framework supports a diverse set of independently trained modalities: spatially-aligned controls such as depth, pose, and edges, camera trajectory with intrinsics, sparse motion control, video editing, and, to our knowledge, the first modular audio-visual controls for a joint generation model. Our method is both compute- and data-efficient: each modality requires only a small dataset and converges within a few hundred to a few thousand training steps, a fraction of the budget of monolithic alternatives. We publicly release our code and trained LoRA checkpoints.</p></description>
|
||||||
<guid isPermaLink="false">https://arxiv.org/abs/2603.24793</guid>
|
<guid isPermaLink="false">https://arxiv.org/abs/2603.24793</guid>
|
||||||
<pubDate>Wed, 25 Mar 2026 20:06:43 +0000</pubDate>
|
<pubDate>Wed, 25 Mar 2026 20:06:43 +0000</pubDate>
|
||||||
</item>
|
</item>
|
||||||
<item>
|
<item>
|
||||||
<title>Calibri: Enhancing Diffusion Transformers via Parameter-Efficient Calibration</title>
|
<title>Calibri: Enhancing Diffusion Transformers via Parameter-Efficient Calibration</title>
|
||||||
<link>https://arxiv.org/abs/2603.24800</link>
|
<link>https://arxiv.org/abs/2603.24800</link>
|
||||||
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.24800.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Danil Tokhchukov, Aysel Mirzoeva, Andrey Kuznetsov, Konstantin Sobolev</p><p><b>Upvotes:</b> 36</p><p><b>Summary:</b> In this paper, we uncover the hidden potential of Diffusion Transformers (DiTs) to significantly enhance generative tasks. Through an in-depth analysis of the denoising process, we demonstrate that introducing a single learned scaling parameter can significantly improve the performance of DiT blocks. Building on this insight, we propose Calibri, a parameter-efficient approach that optimally calibrates DiT components to elevate generative quality. Calibri frames DiT calibration as a black-box reward optimization problem, which is efficiently solved using an evolutionary algorithm and modifies just ~100 parameters. Experimental results reveal that despite its lightweight design, Calibri consistently improves performance across various text-to-image models. Notably, Calibri also reduces the inference steps required for image generation, all while maintaining high-quality outputs.</p></description>
|
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.24800.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Danil Tokhchukov, Aysel Mirzoeva, Andrey Kuznetsov, Konstantin Sobolev</p><p><b>Upvotes:</b> 49</p><p><b>Summary:</b> In this paper, we uncover the hidden potential of Diffusion Transformers (DiTs) to significantly enhance generative tasks. Through an in-depth analysis of the denoising process, we demonstrate that introducing a single learned scaling parameter can significantly improve the performance of DiT blocks. Building on this insight, we propose Calibri, a parameter-efficient approach that optimally calibrates DiT components to elevate generative quality. Calibri frames DiT calibration as a black-box reward optimization problem, which is efficiently solved using an evolutionary algorithm and modifies just ~100 parameters. Experimental results reveal that despite its lightweight design, Calibri consistently improves performance across various text-to-image models. Notably, Calibri also reduces the inference steps required for image generation, all while maintaining high-quality outputs.</p></description>
|
||||||
<guid isPermaLink="false">https://arxiv.org/abs/2603.24800</guid>
|
<guid isPermaLink="false">https://arxiv.org/abs/2603.24800</guid>
|
||||||
<pubDate>Wed, 25 Mar 2026 20:19:50 +0000</pubDate>
|
<pubDate>Wed, 25 Mar 2026 20:19:50 +0000</pubDate>
|
||||||
</item>
|
</item>
|
||||||
<item>
|
<item>
|
||||||
<title>WAFT-Stereo: Warping-Alone Field Transforms for Stereo Matching</title>
|
<title>WAFT-Stereo: Warping-Alone Field Transforms for Stereo Matching</title>
|
||||||
<link>https://arxiv.org/abs/2603.24836</link>
|
<link>https://arxiv.org/abs/2603.24836</link>
|
||||||
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.24836.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yihan Wang, Jia Deng</p><p><b>Upvotes:</b> 0</p><p><b>Summary:</b> We introduce WAFT-Stereo, a simple and effective warping-based method for stereo matching. WAFT-Stereo demonstrates that cost volumes, a common design used in many leading methods, are not necessary for strong performance and can be replaced by warping with improved efficiency. WAFT-Stereo ranks first on ETH3D, KITTI and Middlebury public benchmarks, reducing the zero-shot error by 81% on ETH3D benchmark, while being 1.8-6.7x faster than competitive methods. Code and model weights are available at https://github.com/princeton-vl/WAFT-Stereo.</p></description>
|
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.24836.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yihan Wang, Jia Deng</p><p><b>Upvotes:</b> 1</p><p><b>Summary:</b> We introduce WAFT-Stereo, a simple and effective warping-based method for stereo matching. WAFT-Stereo demonstrates that cost volumes, a common design used in many leading methods, are not necessary for strong performance and can be replaced by warping with improved efficiency. WAFT-Stereo ranks first on ETH3D, KITTI and Middlebury public benchmarks, reducing the zero-shot error by 81% on ETH3D benchmark, while being 1.8-6.7x faster than competitive methods. Code and model weights are available at https://github.com/princeton-vl/WAFT-Stereo.</p></description>
|
||||||
<guid isPermaLink="false">https://arxiv.org/abs/2603.24836</guid>
|
<guid isPermaLink="false">https://arxiv.org/abs/2603.24836</guid>
|
||||||
<pubDate>Wed, 25 Mar 2026 22:07:55 +0000</pubDate>
|
<pubDate>Wed, 25 Mar 2026 22:07:55 +0000</pubDate>
|
||||||
</item>
|
</item>
|
||||||
@@ -116,7 +116,7 @@
|
|||||||
<item>
|
<item>
|
||||||
<title>FinMCP-Bench: Benchmarking LLM Agents for Real-World Financial Tool Use under the Model Context Protocol</title>
|
<title>FinMCP-Bench: Benchmarking LLM Agents for Real-World Financial Tool Use under the Model Context Protocol</title>
|
||||||
<link>https://arxiv.org/abs/2603.24943</link>
|
<link>https://arxiv.org/abs/2603.24943</link>
|
||||||
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.24943.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jie Zhu, Yimin Tian, Boyang Li, Kehao Wu, Zhongzhi Liang, Junhui Li, Xianyin Zhang, Lifan Guo, Feng Chen, Yong Liu, Chi Zhang</p><p><b>Upvotes:</b> 3</p><p><b>Summary:</b> This paper introduces FinMCP-Bench, a novel benchmark for evaluating large language models (LLMs) in solving real-world financial problems through tool invocation of financial model context protocols. FinMCP-Bench contains 613 samples spanning 10 main scenarios and 33 sub-scenarios, featuring both real and synthetic user queries to ensure diversity and authenticity. It incorporates 65 real financial MCPs and three types of samples, single tool, multi-tool, and multi-turn, allowing evaluation of models across different levels of task complexity. Using this benchmark, we systematically assess a range of mainstream LLMs and propose metrics that explicitly measure tool invocation accuracy and reasoning capabilities. FinMCP-Bench provides a standardized, practical, and challenging testbed for advancing research on financial LLM agents.</p></description>
|
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.24943.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jie Zhu, Yimin Tian, Boyang Li, Kehao Wu, Zhongzhi Liang, Junhui Li, Xianyin Zhang, Lifan Guo, Feng Chen, Yong Liu, Chi Zhang</p><p><b>Upvotes:</b> 5</p><p><b>Summary:</b> This paper introduces FinMCP-Bench, a novel benchmark for evaluating large language models (LLMs) in solving real-world financial problems through tool invocation of financial model context protocols. FinMCP-Bench contains 613 samples spanning 10 main scenarios and 33 sub-scenarios, featuring both real and synthetic user queries to ensure diversity and authenticity. It incorporates 65 real financial MCPs and three types of samples, single tool, multi-tool, and multi-turn, allowing evaluation of models across different levels of task complexity. Using this benchmark, we systematically assess a range of mainstream LLMs and propose metrics that explicitly measure tool invocation accuracy and reasoning capabilities. FinMCP-Bench provides a standardized, practical, and challenging testbed for advancing research on financial LLM agents.</p></description>
|
||||||
<guid isPermaLink="false">https://arxiv.org/abs/2603.24943</guid>
|
<guid isPermaLink="false">https://arxiv.org/abs/2603.24943</guid>
|
||||||
<pubDate>Thu, 26 Mar 2026 02:20:04 +0000</pubDate>
|
<pubDate>Thu, 26 Mar 2026 02:20:04 +0000</pubDate>
|
||||||
</item>
|
</item>
|
||||||
@@ -130,7 +130,7 @@
|
|||||||
<item>
|
<item>
|
||||||
<title>Intern-S1-Pro: Scientific Multimodal Foundation Model at Trillion Scale</title>
|
<title>Intern-S1-Pro: Scientific Multimodal Foundation Model at Trillion Scale</title>
|
||||||
<link>https://arxiv.org/abs/2603.25040</link>
|
<link>https://arxiv.org/abs/2603.25040</link>
|
||||||
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.25040.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yicheng Zou, Dongsheng Zhu, Lin Zhu, Tong Zhu, Yunhua Zhou, Peiheng Zhou, Xinyu Zhou, Dongzhan Zhou, Zhiwang Zhou, Yuhao Zhou, Bowen Zhou, Zhanping Zhong, Zhijie Zhong, Haiteng Zhao, Penghao Zhao, Xiaomeng Zhao, Zhiyuan Zhao, Yechen Zhang, Jin Zhang, Wenwei Zhang, Hongjie Zhang, Zhuo Zhang, Wenlong Zhang, Bo Zhang, Chao Zhang, Chen Zhang, Yuhang Zang, Fei Yuan, Jiakang Yuan, Jiashuo Yu, Jinhui Yin, Haochen Ye, Qian Yao, Bowen Yang, Danni Yang, Kaichen Yang, Ziang Yan, Jun Xu, Yicheng Xu, Wanghan Xu, Xuenan Xu, Chao Xu, Ruiliang Xu, Shuhao Xing, Long Xing, Xinchen Xie, Ling-I Wu, Zijian Wu, Zhenyu Wu, Lijun Wu, Yue Wu, Jianyu Wu, Wen Wu, Fan Wu, Xilin Wei, Qi Wei, Bingli Wang, Rui Wang, Ziyi Wang, Zun Wang, Yi Wang, Haomin Wang, Yizhou Wang, Lintao Wang, Yiheng Wang, Longjiang Wang, Bin Wang, Jian Tong, Zhongbo Tian, Huanze Tang, Chen Tang, Shixiang Tang, Yu Sun, Qiushi Sun, Xuerui Su, Qisheng Su, Chenlin Su, Demin Song, Jin Shi, Fukai Shang, Yuchen Ren, Pengli Ren, Xiaoye Qu, Yuan Qu, Jiantao Qiu, Yu Qiao, Runyu Peng, Tianshuo Peng, Jiahui Peng, Qizhi Pei, Zhuoshi Pan, Linke Ouyang, Wenchang Ning, Yichuan Ma, Zerun Ma, Ningsheng Ma, Runyuan Ma, Chengqi Lyu, Haijun Lv, Han Lv, Lindong Lu, Kuikun Liu, Jiangning Liu, Yuhong Liu, Kai Liu, Hongwei Liu, Zhoumianze Liu, Mengjie Liu, Ziyu Liu, Wenran Liu, Yang Liu, Liwei Liu, Kaiwen Liu, Junyao Lin, Junming Lin, Tianyang Lin, Dahua Lin, Jianze Liang, Linyang Li, Peiji Li, Zonglin Li, Zehao Li, Pengze Li, Guoyan Li, Lingkai Kong, Linglin Jing, Zhenjiang Jin, Feifei Jiang, Qian Jiang, Junhao Huang, Zixian Huang, Haian Huang, Zhouqi Hua, Han Hu, Linfeng Hou, Yinan He, Conghui He, Tianyao He, Xu Guo, Qipeng Guo, Aijia Guo, Yuzhe Gu, Lixin Gu, Jingyang Gong, Qiming Ge, Jiaye Ge, Songyang Gao, Jianfei Gao, Xinyu Fang, Caihua fan, Yue Fan, Yanhui Duan, Zichen Ding, Shengyuan Ding, Xuanlang Dai, Erfei Cui, Ganqu Cui, Pei Chu, Tao Chu, Guangran Cheng, Yu Cheng, Kai Chen, Yongkang Chen, Chiyu Chen, Guanzhou Chen, Qiaosheng Chen, Sitao Chen, Xin Chen, Haojiong Chen, Yicheng Chen, Weihan Cao, Yuhang Cao, Qinglong Cao, Lei Bai</p><p><b>Upvotes:</b> 86</p><p><b>Summary:</b> We introduce Intern-S1-Pro, the first one-trillion-parameter scientific multimodal foundation model. Scaling to this unprecedented size, the model delivers a comprehensive enhancement across both general and scientific domains. Beyond stronger reasoning and image-text understanding capabilities, its intelligence is augmented with advanced agent capabilities. Simultaneously, its scientific expertise has been vastly expanded to master over 100 specialized tasks across critical science fields, including chemistry, materials, life sciences, and earth sciences. Achieving this massive scale is made possible by the robust infrastructure support of XTuner and LMDeploy, which facilitates highly efficient Reinforcement Learning (RL) training at the 1-trillion parameter level while ensuring strict precision consistency between training and inference. By seamlessly integrating these advancements, Intern-S1-Pro further fortifies the fusion of general and specialized intelligence, working as a Specializable Generalist, demonstrating its position in the top tier of open-source models for general capabilities, while outperforming proprietary models in the depth of specialized scientific tasks.</p></description>
|
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.25040.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yicheng Zou, Dongsheng Zhu, Lin Zhu, Tong Zhu, Yunhua Zhou, Peiheng Zhou, Xinyu Zhou, Dongzhan Zhou, Zhiwang Zhou, Yuhao Zhou, Bowen Zhou, Zhanping Zhong, Zhijie Zhong, Haiteng Zhao, Penghao Zhao, Xiaomeng Zhao, Zhiyuan Zhao, Yechen Zhang, Jin Zhang, Wenwei Zhang, Hongjie Zhang, Zhuo Zhang, Wenlong Zhang, Bo Zhang, Chao Zhang, Chen Zhang, Yuhang Zang, Fei Yuan, Jiakang Yuan, Jiashuo Yu, Jinhui Yin, Haochen Ye, Qian Yao, Bowen Yang, Danni Yang, Kaichen Yang, Ziang Yan, Jun Xu, Yicheng Xu, Wanghan Xu, Xuenan Xu, Chao Xu, Ruiliang Xu, Shuhao Xing, Long Xing, Xinchen Xie, Ling-I Wu, Zijian Wu, Zhenyu Wu, Lijun Wu, Yue Wu, Jianyu Wu, Wen Wu, Fan Wu, Xilin Wei, Qi Wei, Bingli Wang, Rui Wang, Ziyi Wang, Zun Wang, Yi Wang, Haomin Wang, Yizhou Wang, Lintao Wang, Yiheng Wang, Longjiang Wang, Bin Wang, Jian Tong, Zhongbo Tian, Huanze Tang, Chen Tang, Shixiang Tang, Yu Sun, Qiushi Sun, Xuerui Su, Qisheng Su, Chenlin Su, Demin Song, Jin Shi, Fukai Shang, Yuchen Ren, Pengli Ren, Xiaoye Qu, Yuan Qu, Jiantao Qiu, Yu Qiao, Runyu Peng, Tianshuo Peng, Jiahui Peng, Qizhi Pei, Zhuoshi Pan, Linke Ouyang, Wenchang Ning, Yichuan Ma, Zerun Ma, Ningsheng Ma, Runyuan Ma, Chengqi Lyu, Haijun Lv, Han Lv, Lindong Lu, Kuikun Liu, Jiangning Liu, Yuhong Liu, Kai Liu, Hongwei Liu, Zhoumianze Liu, Mengjie Liu, Ziyu Liu, Wenran Liu, Yang Liu, Liwei Liu, Kaiwen Liu, Junyao Lin, Junming Lin, Tianyang Lin, Dahua Lin, Jianze Liang, Linyang Li, Peiji Li, Zonglin Li, Zehao Li, Pengze Li, Guoyan Li, Lingkai Kong, Linglin Jing, Zhenjiang Jin, Feifei Jiang, Qian Jiang, Junhao Huang, Zixian Huang, Haian Huang, Zhouqi Hua, Han Hu, Linfeng Hou, Yinan He, Conghui He, Tianyao He, Xu Guo, Qipeng Guo, Aijia Guo, Yuzhe Gu, Lixin Gu, Jingyang Gong, Qiming Ge, Jiaye Ge, Songyang Gao, Jianfei Gao, Xinyu Fang, Caihua fan, Yue Fan, Yanhui Duan, Zichen Ding, Shengyuan Ding, Xuanlang Dai, Erfei Cui, Ganqu Cui, Pei Chu, Tao Chu, Guangran Cheng, Yu Cheng, Kai Chen, Yongkang Chen, Chiyu Chen, Guanzhou Chen, Qiaosheng Chen, Sitao Chen, Xin Chen, Haojiong Chen, Yicheng Chen, Weihan Cao, Yuhang Cao, Qinglong Cao, Lei Bai</p><p><b>Upvotes:</b> 100</p><p><b>Summary:</b> We introduce Intern-S1-Pro, the first one-trillion-parameter scientific multimodal foundation model. Scaling to this unprecedented size, the model delivers a comprehensive enhancement across both general and scientific domains. Beyond stronger reasoning and image-text understanding capabilities, its intelligence is augmented with advanced agent capabilities. Simultaneously, its scientific expertise has been vastly expanded to master over 100 specialized tasks across critical science fields, including chemistry, materials, life sciences, and earth sciences. Achieving this massive scale is made possible by the robust infrastructure support of XTuner and LMDeploy, which facilitates highly efficient Reinforcement Learning (RL) training at the 1-trillion parameter level while ensuring strict precision consistency between training and inference. By seamlessly integrating these advancements, Intern-S1-Pro further fortifies the fusion of general and specialized intelligence, working as a Specializable Generalist, demonstrating its position in the top tier of open-source models for general capabilities, while outperforming proprietary models in the depth of specialized scientific tasks.</p></description>
|
||||||
<guid isPermaLink="false">https://arxiv.org/abs/2603.25040</guid>
|
<guid isPermaLink="false">https://arxiv.org/abs/2603.25040</guid>
|
||||||
<pubDate>Thu, 26 Mar 2026 05:21:45 +0000</pubDate>
|
<pubDate>Thu, 26 Mar 2026 05:21:45 +0000</pubDate>
|
||||||
</item>
|
</item>
|
||||||
@@ -151,21 +151,21 @@
|
|||||||
<item>
|
<item>
|
||||||
<title>RealRestorer: Towards Generalizable Real-World Image Restoration with Large-Scale Image Editing Models</title>
|
<title>RealRestorer: Towards Generalizable Real-World Image Restoration with Large-Scale Image Editing Models</title>
|
||||||
<link>https://arxiv.org/abs/2603.25502</link>
|
<link>https://arxiv.org/abs/2603.25502</link>
|
||||||
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.25502.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yufeng Yang, Xianfang Zeng, Zhangqi Jiang, Fukun Yin, Jianzhuang Liu, Wei Cheng, jinghong lan, Shiyu Liu, Yuqi Peng, Gang YU, Shifeng Chen</p><p><b>Upvotes:</b> 37</p><p><b>Summary:</b> Image restoration under real-world degradations is critical for downstream tasks such as autonomous driving and object detection. However, existing restoration models are often limited by the scale and distribution of their training data, resulting in poor generalization to real-world scenarios. Recently, large-scale image editing models have shown strong generalization ability in restoration tasks, especially for closed-source models like Nano Banana Pro, which can restore images while preserving consistency. Nevertheless, achieving such performance with those large universal models requires substantial data and computational costs. To address this issue, we construct a large-scale dataset covering nine common real-world degradation types and train a state-of-the-art open-source model to narrow the gap with closed-source alternatives. Furthermore, we introduce RealIR-Bench, which contains 464 real-world degraded images and tailored evaluation metrics focusing on degradation removal and consistency preservation. Extensive experiments demonstrate our model ranks first among open-source methods, achieving state-of-the-art performance.</p></description>
|
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.25502.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yufeng Yang, Xianfang Zeng, Zhangqi Jiang, Fukun Yin, Jianzhuang Liu, Wei Cheng, jinghong lan, Shiyu Liu, Yuqi Peng, Gang YU, Shifeng Chen</p><p><b>Upvotes:</b> 43</p><p><b>Summary:</b> Image restoration under real-world degradations is critical for downstream tasks such as autonomous driving and object detection. However, existing restoration models are often limited by the scale and distribution of their training data, resulting in poor generalization to real-world scenarios. Recently, large-scale image editing models have shown strong generalization ability in restoration tasks, especially for closed-source models like Nano Banana Pro, which can restore images while preserving consistency. Nevertheless, achieving such performance with those large universal models requires substantial data and computational costs. To address this issue, we construct a large-scale dataset covering nine common real-world degradation types and train a state-of-the-art open-source model to narrow the gap with closed-source alternatives. Furthermore, we introduce RealIR-Bench, which contains 464 real-world degraded images and tailored evaluation metrics focusing on degradation removal and consistency preservation. Extensive experiments demonstrate our model ranks first among open-source methods, achieving state-of-the-art performance.</p></description>
|
||||||
<guid isPermaLink="false">https://arxiv.org/abs/2603.25502</guid>
|
<guid isPermaLink="false">https://arxiv.org/abs/2603.25502</guid>
|
||||||
<pubDate>Thu, 26 Mar 2026 14:39:39 +0000</pubDate>
|
<pubDate>Thu, 26 Mar 2026 14:39:39 +0000</pubDate>
|
||||||
</item>
|
</item>
|
||||||
<item>
|
<item>
|
||||||
<title>Voxtral TTS</title>
|
<title>Voxtral TTS</title>
|
||||||
<link>https://arxiv.org/abs/2603.25551</link>
|
<link>https://arxiv.org/abs/2603.25551</link>
|
||||||
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.25551.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Alexander H. Liu, Alexis Tacnet, Andy Ehrenberg, Andy Lo, Chen-Yo Sun, Guillaume Lample, Henry Lagarde, Jean-Malo Delignon, Jaeyoung Kim, John Harvill, Khyathi Raghavi Chandu, Lorenzo Signoretti, Margaret Jennings, Patrick von Platen, Pavankumar Reddy Muddireddy, Rohin Arora, Sanchit Gandhi, Samuel Humeau, Soham Ghosh, Srijan Mishra, Van Phung, Abdelaziz Bounhar, Abhinav Rastogi, Adrien Sadé, Alan Jeffares, Albert Jiang, Alexandre Cahill, Alexandre Gavaudan, Alexandre Sablayrolles, Amélie Héliou, Amos You, Andrew Bai, Andrew Zhao, Angele Lenglemetz, Anmol Agarwal, Anton Eliseev, Antonia Calvi, Arjun Majumdar, Arthur Fournier, Artjom Joosen, Avi Sooriyarachchi, Aysenur Karaduman Utkur, Baptiste Bout, Baptiste Rozière, Baudouin De Monicault, Benjamin Tibi, Bowen Yang, Charlotte Cronjäger, Clémence Lanfranchi, Connor Chen, Corentin Barreau, Corentin Sautier, Cyprien Courtot, Darius Dabert, Diego de las Casas, Elizaveta Demyanenko, Elliot Chane-Sane, Emmanuel Gottlob, Enguerrand Paquin, Etienne Goffinet, Fabien Niel, Faruk Ahmed, Federico Baldassarre, Gabrielle Berrada, Gaëtan Ecrepont, Gauthier Guinet, Genevieve Hayes, Georgii Novikov, Giada Pistilli, Guillaume Kunsch, Guillaume Martin, Guillaume Raille, Gunjan Dhanuka, Gunshi Gupta, Han Zhou, Harshil Shah, Hope McGovern, Hugo Thimonier, Indraneel Mukherjee, Irene Zhang, Jacques Sun, Jan Ludziejewski, Jason Rute, Jérémie Dentan, Joachim Studnia, Jonas Amar, Joséphine Delas, Josselin Somerville Roberts, Julien Tauran, Karmesh Yadav, Kartik Khandelwal, Kilian Tep, Kush Jain, Laurence Aitchison, Laurent Fainsin, Léonard Blier, Lingxiao Zhao, Louis Martin, Lucile Saulnier, Luyu Gao, Maarten Buyl, Manan Sharma, Marie Pellat, Mark Prins, Martin Alexandre, Mathieu Poirée, Mathieu Schmitt, Mathilde Guillaumin, Matthieu Dinot, Matthieu Futeral, Maxime Darrin, Maximilian Augustin, Mert Unsal, Mia Chiquier, Mikhail Biriuchinskii, Minh-Quang Pham, Mircea Lica, Morgane Rivière, Nathan Grinsztajn, Neha Gupta, Olivier Bousquet, Olivier Duchenne, Patricia Wang, Paul Jacob, Paul Wambergue, Paula Kurylowicz, Philippe Pinel, Philomène Chagniot, Pierre Stock, Piotr Miłoś, Prateek Gupta, Pravesh Agrawal, Quentin Torroba, Ram Ramrakhya, Randall Isenhour, Rishi Shah, Romain Sauvestre, Roman Soletskyi, Rosalie Millner, Rupert Menneer, Sagar Vaze, Samuel Barry, Samuel Belkadi, Sandeep Subramanian, Sean Cha, Shashwat Verma, Siddhant Waghjale, Siddharth Gandhi, Simon Lepage, Sumukh Aithal, Szymon Antoniak, Tarun Kumar Vangani, Teven Le Scao, Théo Cachet, Theo Simon Sorg, Thibaut Lavril, Thomas Chabal, Thomas Foubert, Thomas Robert, Thomas Wang, Tim Lawson, Tom Bewley, Tom Edwards, Tyler Wang, Umar Jamil, Umberto Tomasini, Valeriia Nemychnikova, Vedant Nanda, Victor Jouault, Vincent Maladière, Vincent Pfister, Virgile Richard, Vladislav Bataev, Wassim Bouaziz, Wen-Ding Li, William Havard, William Marshall, Xinghui Li, Xingran Guo, Xinyu Yang, Yannic Neuhaus, Yassine El Ouahidi, Yassir Bendou, Yihan Wang, Yimu Pan, Zaccharie Ramzi, Zhenlin Xu</p><p><b>Upvotes:</b> 23</p><p><b>Summary:</b> We introduce Voxtral TTS, an expressive multilingual text-to-speech model that generates natural speech from as little as 3 seconds of reference audio. Voxtral TTS adopts a hybrid architecture that combines auto-regressive generation of semantic speech tokens with flow-matching for acoustic tokens. These tokens are encoded and decoded with Voxtral Codec, a speech tokenizer trained from scratch with a hybrid VQ-FSQ quantization scheme. In human evaluations conducted by native speakers, Voxtral TTS is preferred for multilingual voice cloning due to its naturalness and expressivity, achieving a 68.4\% win rate over ElevenLabs Flash v2.5. We release the model weights under a CC BY-NC license.</p></description>
|
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.25551.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Alexander H. Liu, Alexis Tacnet, Andy Ehrenberg, Andy Lo, Chen-Yo Sun, Guillaume Lample, Henry Lagarde, Jean-Malo Delignon, Jaeyoung Kim, John Harvill, Khyathi Raghavi Chandu, Lorenzo Signoretti, Margaret Jennings, Patrick von Platen, Pavankumar Reddy Muddireddy, Rohin Arora, Sanchit Gandhi, Samuel Humeau, Soham Ghosh, Srijan Mishra, Van Phung, Abdelaziz Bounhar, Abhinav Rastogi, Adrien Sadé, Alan Jeffares, Albert Jiang, Alexandre Cahill, Alexandre Gavaudan, Alexandre Sablayrolles, Amélie Héliou, Amos You, Andrew Bai, Andrew Zhao, Angele Lenglemetz, Anmol Agarwal, Anton Eliseev, Antonia Calvi, Arjun Majumdar, Arthur Fournier, Artjom Joosen, Avi Sooriyarachchi, Aysenur Karaduman Utkur, Baptiste Bout, Baptiste Rozière, Baudouin De Monicault, Benjamin Tibi, Bowen Yang, Charlotte Cronjäger, Clémence Lanfranchi, Connor Chen, Corentin Barreau, Corentin Sautier, Cyprien Courtot, Darius Dabert, Diego de las Casas, Elizaveta Demyanenko, Elliot Chane-Sane, Emmanuel Gottlob, Enguerrand Paquin, Etienne Goffinet, Fabien Niel, Faruk Ahmed, Federico Baldassarre, Gabrielle Berrada, Gaëtan Ecrepont, Gauthier Guinet, Genevieve Hayes, Georgii Novikov, Giada Pistilli, Guillaume Kunsch, Guillaume Martin, Guillaume Raille, Gunjan Dhanuka, Gunshi Gupta, Han Zhou, Harshil Shah, Hope McGovern, Hugo Thimonier, Indraneel Mukherjee, Irene Zhang, Jacques Sun, Jan Ludziejewski, Jason Rute, Jérémie Dentan, Joachim Studnia, Jonas Amar, Joséphine Delas, Josselin Somerville Roberts, Julien Tauran, Karmesh Yadav, Kartik Khandelwal, Kilian Tep, Kush Jain, Laurence Aitchison, Laurent Fainsin, Léonard Blier, Lingxiao Zhao, Louis Martin, Lucile Saulnier, Luyu Gao, Maarten Buyl, Manan Sharma, Marie Pellat, Mark Prins, Martin Alexandre, Mathieu Poirée, Mathieu Schmitt, Mathilde Guillaumin, Matthieu Dinot, Matthieu Futeral, Maxime Darrin, Maximilian Augustin, Mert Unsal, Mia Chiquier, Mikhail Biriuchinskii, Minh-Quang Pham, Mircea Lica, Morgane Rivière, Nathan Grinsztajn, Neha Gupta, Olivier Bousquet, Olivier Duchenne, Patricia Wang, Paul Jacob, Paul Wambergue, Paula Kurylowicz, Philippe Pinel, Philomène Chagniot, Pierre Stock, Piotr Miłoś, Prateek Gupta, Pravesh Agrawal, Quentin Torroba, Ram Ramrakhya, Randall Isenhour, Rishi Shah, Romain Sauvestre, Roman Soletskyi, Rosalie Millner, Rupert Menneer, Sagar Vaze, Samuel Barry, Samuel Belkadi, Sandeep Subramanian, Sean Cha, Shashwat Verma, Siddhant Waghjale, Siddharth Gandhi, Simon Lepage, Sumukh Aithal, Szymon Antoniak, Tarun Kumar Vangani, Teven Le Scao, Théo Cachet, Theo Simon Sorg, Thibaut Lavril, Thomas Chabal, Thomas Foubert, Thomas Robert, Thomas Wang, Tim Lawson, Tom Bewley, Tom Edwards, Tyler Wang, Umar Jamil, Umberto Tomasini, Valeriia Nemychnikova, Vedant Nanda, Victor Jouault, Vincent Maladière, Vincent Pfister, Virgile Richard, Vladislav Bataev, Wassim Bouaziz, Wen-Ding Li, William Havard, William Marshall, Xinghui Li, Xingran Guo, Xinyu Yang, Yannic Neuhaus, Yassine El Ouahidi, Yassir Bendou, Yihan Wang, Yimu Pan, Zaccharie Ramzi, Zhenlin Xu</p><p><b>Upvotes:</b> 30</p><p><b>Summary:</b> We introduce Voxtral TTS, an expressive multilingual text-to-speech model that generates natural speech from as little as 3 seconds of reference audio. Voxtral TTS adopts a hybrid architecture that combines auto-regressive generation of semantic speech tokens with flow-matching for acoustic tokens. These tokens are encoded and decoded with Voxtral Codec, a speech tokenizer trained from scratch with a hybrid VQ-FSQ quantization scheme. In human evaluations conducted by native speakers, Voxtral TTS is preferred for multilingual voice cloning due to its naturalness and expressivity, achieving a 68.4\% win rate over ElevenLabs Flash v2.5. We release the model weights under a CC BY-NC license.</p></description>
|
||||||
<guid isPermaLink="false">https://arxiv.org/abs/2603.25551</guid>
|
<guid isPermaLink="false">https://arxiv.org/abs/2603.25551</guid>
|
||||||
<pubDate>Thu, 26 Mar 2026 15:23:34 +0000</pubDate>
|
<pubDate>Thu, 26 Mar 2026 15:23:34 +0000</pubDate>
|
||||||
</item>
|
</item>
|
||||||
<item>
|
<item>
|
||||||
<title>Revisiting On-Policy Distillation: Empirical Failure Modes and Simple Fixes</title>
|
<title>Revisiting On-Policy Distillation: Empirical Failure Modes and Simple Fixes</title>
|
||||||
<link>https://arxiv.org/abs/2603.25562</link>
|
<link>https://arxiv.org/abs/2603.25562</link>
|
||||||
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.25562.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yuqian Fu, Haohuan Huang, Kaiwen Jiang, Yuanheng Zhu, Dongbin Zhao</p><p><b>Upvotes:</b> 2</p><p><b>Summary:</b> On-policy distillation (OPD) is appealing for large language model (LLM) post-training because it evaluates teacher feedback on student-generated rollouts rather than fixed teacher traces. In long-horizon settings, however, the common sampled-token variant is fragile: it reduces distribution matching to a one-token signal and becomes increasingly unreliable as rollouts drift away from prefixes the teacher commonly visits. We revisit OPD from the estimator and implementation sides. Theoretically, token-level OPD is biased relative to sequence-level reverse-KL, but it has a much tighter worst-case variance bound; our toy study shows the same tradeoff empirically, with stronger future-reward coupling producing higher gradient variance and less stable learning. Empirically, we identify three failure modes of sampled-token OPD: an imbalanced one-token signal, unreliable teacher guidance on student-generated prefixes, and distortions caused by tokenizer or special-token mismatch. We address these issues with teacher top-K local support matching, implemented as truncated reverse-KL with top-p rollout sampling and special-token masking. Across single-task math reasoning and multi-task agentic-plus-math training, this objective yields more stable optimization and better downstream performance than sampled-token OPD.</p></description>
|
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.25562.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yuqian Fu, Haohuan Huang, Kaiwen Jiang, Yuanheng Zhu, Dongbin Zhao</p><p><b>Upvotes:</b> 3</p><p><b>Summary:</b> On-policy distillation (OPD) is appealing for large language model (LLM) post-training because it evaluates teacher feedback on student-generated rollouts rather than fixed teacher traces. In long-horizon settings, however, the common sampled-token variant is fragile: it reduces distribution matching to a one-token signal and becomes increasingly unreliable as rollouts drift away from prefixes the teacher commonly visits. We revisit OPD from the estimator and implementation sides. Theoretically, token-level OPD is biased relative to sequence-level reverse-KL, but it has a much tighter worst-case variance bound; our toy study shows the same tradeoff empirically, with stronger future-reward coupling producing higher gradient variance and less stable learning. Empirically, we identify three failure modes of sampled-token OPD: an imbalanced one-token signal, unreliable teacher guidance on student-generated prefixes, and distortions caused by tokenizer or special-token mismatch. We address these issues with teacher top-K local support matching, implemented as truncated reverse-KL with top-p rollout sampling and special-token masking. Across single-task math reasoning and multi-task agentic-plus-math training, this objective yields more stable optimization and better downstream performance than sampled-token OPD.</p></description>
|
||||||
<guid isPermaLink="false">https://arxiv.org/abs/2603.25562</guid>
|
<guid isPermaLink="false">https://arxiv.org/abs/2603.25562</guid>
|
||||||
<pubDate>Thu, 26 Mar 2026 15:35:59 +0000</pubDate>
|
<pubDate>Thu, 26 Mar 2026 15:35:59 +0000</pubDate>
|
||||||
</item>
|
</item>
|
||||||
@@ -186,28 +186,28 @@
|
|||||||
<item>
|
<item>
|
||||||
<title>PixelSmile: Toward Fine-Grained Facial Expression Editing</title>
|
<title>PixelSmile: Toward Fine-Grained Facial Expression Editing</title>
|
||||||
<link>https://arxiv.org/abs/2603.25728</link>
|
<link>https://arxiv.org/abs/2603.25728</link>
|
||||||
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.25728.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jiabin Hua, Hengyuan Xu, Aojie Li, Wei Cheng, Gang Yu, Xingjun Ma, Yu-Gang Jiang</p><p><b>Upvotes:</b> 97</p><p><b>Summary:</b> Fine-grained facial expression editing has long been limited by intrinsic semantic overlap. To address this, we construct the Flex Facial Expression (FFE) dataset with continuous affective annotations and establish FFE-Bench to evaluate structural confusion, editing accuracy, linear controllability, and the trade-off between expression editing and identity preservation. We propose PixelSmile, a diffusion framework that disentangles expression semantics via fully symmetric joint training. PixelSmile combines intensity supervision with contrastive learning to produce stronger and more distinguishable expressions, achieving precise and stable linear expression control through textual latent interpolation. Extensive experiments demonstrate that PixelSmile achieves superior disentanglement and robust identity preservation, confirming its effectiveness for continuous, controllable, and fine-grained expression editing, while naturally supporting smooth expression blending.</p></description>
|
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.25728.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Jiabin Hua, Hengyuan Xu, Aojie Li, Wei Cheng, Gang Yu, Xingjun Ma, Yu-Gang Jiang</p><p><b>Upvotes:</b> 105</p><p><b>Summary:</b> Fine-grained facial expression editing has long been limited by intrinsic semantic overlap. To address this, we construct the Flex Facial Expression (FFE) dataset with continuous affective annotations and establish FFE-Bench to evaluate structural confusion, editing accuracy, linear controllability, and the trade-off between expression editing and identity preservation. We propose PixelSmile, a diffusion framework that disentangles expression semantics via fully symmetric joint training. PixelSmile combines intensity supervision with contrastive learning to produce stronger and more distinguishable expressions, achieving precise and stable linear expression control through textual latent interpolation. Extensive experiments demonstrate that PixelSmile achieves superior disentanglement and robust identity preservation, confirming its effectiveness for continuous, controllable, and fine-grained expression editing, while naturally supporting smooth expression blending.</p></description>
|
||||||
<guid isPermaLink="false">https://arxiv.org/abs/2603.25728</guid>
|
<guid isPermaLink="false">https://arxiv.org/abs/2603.25728</guid>
|
||||||
<pubDate>Thu, 26 Mar 2026 17:59:04 +0000</pubDate>
|
<pubDate>Thu, 26 Mar 2026 17:59:04 +0000</pubDate>
|
||||||
</item>
|
</item>
|
||||||
<item>
|
<item>
|
||||||
<title>Vega: Learning to Drive with Natural Language Instructions</title>
|
<title>Vega: Learning to Drive with Natural Language Instructions</title>
|
||||||
<link>https://arxiv.org/abs/2603.25741</link>
|
<link>https://arxiv.org/abs/2603.25741</link>
|
||||||
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.25741.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Sicheng Zuo, Yuxuan Li, Wenzhao Zheng, Zheng Zhu, Jie Zhou, Jiwen Lu</p><p><b>Upvotes:</b> 0</p><p><b>Summary:</b> Vision-language-action models have reshaped autonomous driving to incorporate languages into the decision-making process. However, most existing pipelines only utilize the language modality for scene descriptions or reasoning and lack the flexibility to follow diverse user instructions for personalized driving. To address this, we first construct a large-scale driving dataset (InstructScene) containing around 100,000 scenes annotated with diverse driving instructions with the corresponding trajectories. We then propose a unified Vision-Language-World-Action model, Vega, for instruction-based generation and planning. We employ the autoregressive paradigm to process visual inputs (vision) and language instructions (language) and the diffusion paradigm to generate future predictions (world modeling) and trajectories (action). We perform joint attention to enable interactions between the modalities and use individual projection layers for different modalities for more capabilities. Extensive experiments demonstrate that our method not only achieves superior planning performance but also exhibits strong instruction-following abilities, paving the way for more intelligent and personalized driving systems.</p></description>
|
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.25741.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Sicheng Zuo, Yuxuan Li, Wenzhao Zheng, Zheng Zhu, Jie Zhou, Jiwen Lu</p><p><b>Upvotes:</b> 4</p><p><b>Summary:</b> Vision-language-action models have reshaped autonomous driving to incorporate languages into the decision-making process. However, most existing pipelines only utilize the language modality for scene descriptions or reasoning and lack the flexibility to follow diverse user instructions for personalized driving. To address this, we first construct a large-scale driving dataset (InstructScene) containing around 100,000 scenes annotated with diverse driving instructions with the corresponding trajectories. We then propose a unified Vision-Language-World-Action model, Vega, for instruction-based generation and planning. We employ the autoregressive paradigm to process visual inputs (vision) and language instructions (language) and the diffusion paradigm to generate future predictions (world modeling) and trajectories (action). We perform joint attention to enable interactions between the modalities and use individual projection layers for different modalities for more capabilities. Extensive experiments demonstrate that our method not only achieves superior planning performance but also exhibits strong instruction-following abilities, paving the way for more intelligent and personalized driving systems.</p></description>
|
||||||
<guid isPermaLink="false">https://arxiv.org/abs/2603.25741</guid>
|
<guid isPermaLink="false">https://arxiv.org/abs/2603.25741</guid>
|
||||||
<pubDate>Thu, 26 Mar 2026 17:59:56 +0000</pubDate>
|
<pubDate>Thu, 26 Mar 2026 17:59:56 +0000</pubDate>
|
||||||
</item>
|
</item>
|
||||||
<item>
|
<item>
|
||||||
<title>MuRF: Unlocking the Multi-Scale Potential of Vision Foundation Models</title>
|
<title>MuRF: Unlocking the Multi-Scale Potential of Vision Foundation Models</title>
|
||||||
<link>https://arxiv.org/abs/2603.25744</link>
|
<link>https://arxiv.org/abs/2603.25744</link>
|
||||||
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.25744.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Bocheng Zou, Mu Cai, Mark Stanley, Dingfu Lu, Yong Jae Lee</p><p><b>Upvotes:</b> 4</p><p><b>Summary:</b> Vision Foundation Models (VFMs) have become the cornerstone of modern computer vision, offering robust representations across a wide array of tasks. While recent advances allow these models to handle varying input sizes during training, inference typically remains restricted to a single, fixed scale. This prevalent single-scale paradigm overlooks a fundamental property of visual perception: varying resolutions offer complementary inductive biases, where low-resolution views excel at global semantic recognition and high-resolution views are essential for fine-grained refinement. In this work, we propose Multi-Resolution Fusion (MuRF), a simple yet universally effective strategy to harness this synergy at inference time. Instead of relying on a single view, MuRF constructs a unified representation by processing an image at multiple resolutions through a frozen VFM and fusing the resulting features. The universality of MuRF is its most compelling attribute. It is not tied to a specific architecture, serving instead as a fundamental, training-free enhancement to visual representation. We empirically validate this by applying MuRF to a broad spectrum of critical computer vision tasks across multiple distinct VFM families - primarily DINOv2, but also demonstrating successful generalization to contrastive models like SigLIP2.</p></description>
|
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.25744.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Bocheng Zou, Mu Cai, Mark Stanley, Dingfu Lu, Yong Jae Lee</p><p><b>Upvotes:</b> 8</p><p><b>Summary:</b> Vision Foundation Models (VFMs) have become the cornerstone of modern computer vision, offering robust representations across a wide array of tasks. While recent advances allow these models to handle varying input sizes during training, inference typically remains restricted to a single, fixed scale. This prevalent single-scale paradigm overlooks a fundamental property of visual perception: varying resolutions offer complementary inductive biases, where low-resolution views excel at global semantic recognition and high-resolution views are essential for fine-grained refinement. In this work, we propose Multi-Resolution Fusion (MuRF), a simple yet universally effective strategy to harness this synergy at inference time. Instead of relying on a single view, MuRF constructs a unified representation by processing an image at multiple resolutions through a frozen VFM and fusing the resulting features. The universality of MuRF is its most compelling attribute. It is not tied to a specific architecture, serving instead as a fundamental, training-free enhancement to visual representation. We empirically validate this by applying MuRF to a broad spectrum of critical computer vision tasks across multiple distinct VFM families - primarily DINOv2, but also demonstrating successful generalization to contrastive models like SigLIP2.</p></description>
|
||||||
<guid isPermaLink="false">https://arxiv.org/abs/2603.25744</guid>
|
<guid isPermaLink="false">https://arxiv.org/abs/2603.25744</guid>
|
||||||
<pubDate>Thu, 26 Mar 2026 17:59:58 +0000</pubDate>
|
<pubDate>Thu, 26 Mar 2026 17:59:58 +0000</pubDate>
|
||||||
</item>
|
</item>
|
||||||
<item>
|
<item>
|
||||||
<title>Less Gaussians, Texture More: 4K Feed-Forward Textured Splatting</title>
|
<title>Less Gaussians, Texture More: 4K Feed-Forward Textured Splatting</title>
|
||||||
<link>https://arxiv.org/abs/2603.25745</link>
|
<link>https://arxiv.org/abs/2603.25745</link>
|
||||||
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.25745.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yixing Lao, Xuyang Bai, Xiaoyang Wu, Nuoyuan Yan, Zixin Luo, Tian Fang, Jean-Daniel Nahmias, Yanghai Tsin, Shiwei Li, Hengshuang Zhao</p><p><b>Upvotes:</b> 5</p><p><b>Summary:</b> Existing feed-forward 3D Gaussian Splatting methods predict pixel-aligned primitives, leading to a quadratic growth in primitive count as resolution increases. This fundamentally limits their scalability, making high-resolution synthesis such as 4K intractable. We introduce LGTM (Less Gaussians, Texture More), a feed-forward framework that overcomes this resolution scaling barrier. By predicting compact Gaussian primitives coupled with per-primitive textures, LGTM decouples geometric complexity from rendering resolution. This approach enables high-fidelity 4K novel view synthesis without per-scene optimization, a capability previously out of reach for feed-forward methods, all while using significantly fewer Gaussian primitives. Project page: https://yxlao.github.io/lgtm/</p></description>
|
<description><p><img src="https://cdn-thumbnails.huggingface.co/social-thumbnails/papers/2603.25745.png" alt="Paper thumbnail" style="max-width: 300px; height: auto;" /></p><p><b>Authors:</b> Yixing Lao, Xuyang Bai, Xiaoyang Wu, Nuoyuan Yan, Zixin Luo, Tian Fang, Jean-Daniel Nahmias, Yanghai Tsin, Shiwei Li, Hengshuang Zhao</p><p><b>Upvotes:</b> 8</p><p><b>Summary:</b> Existing feed-forward 3D Gaussian Splatting methods predict pixel-aligned primitives, leading to a quadratic growth in primitive count as resolution increases. This fundamentally limits their scalability, making high-resolution synthesis such as 4K intractable. We introduce LGTM (Less Gaussians, Texture More), a feed-forward framework that overcomes this resolution scaling barrier. By predicting compact Gaussian primitives coupled with per-primitive textures, LGTM decouples geometric complexity from rendering resolution. This approach enables high-fidelity 4K novel view synthesis without per-scene optimization, a capability previously out of reach for feed-forward methods, all while using significantly fewer Gaussian primitives. Project page: https://yxlao.github.io/lgtm/</p></description>
|
||||||
<guid isPermaLink="false">https://arxiv.org/abs/2603.25745</guid>
|
<guid isPermaLink="false">https://arxiv.org/abs/2603.25745</guid>
|
||||||
<pubDate>Thu, 26 Mar 2026 17:59:59 +0000</pubDate>
|
<pubDate>Thu, 26 Mar 2026 17:59:59 +0000</pubDate>
|
||||||
</item>
|
</item>
|
||||||
|
|||||||
Reference in New Issue
Block a user