Daily Papers of 2025-07-02

  1. GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning 176 upvotes, #1 of 2025-07-02
  2. Does Math Reasoning Improve General LLM Capabilities? Understanding Transferability of LLM Reasoning 62 upvotes, #2 of 2025-07-02
  3. SciArena: An Open Evaluation Platform for Foundation Models in Scientific Literature Tasks 42 upvotes, #3 of 2025-07-02
  4. Radial Attention: O(nlog n) Sparse Attention with Energy Decay for Long Video Generation 36 upvotes, #4 of 2025-07-02
  5. MoCa: Modality-aware Continual Pre-training Makes Better Bidirectional Multimodal Embeddings 36 upvotes, #4 of 2025-07-02
  6. DiffuCoder: Understanding and Improving Masked Diffusion Models for Code Generation 26 upvotes, #6 of 2025-07-02
  7. Thinking Beyond Tokens: From Brain-Inspired Intelligence to Cognitive Foundations for Artificial General Intelligence and its Societal Impact 20 upvotes, #7 of 2025-07-02
  8. HumanOmniV2: From Understanding to Omni-Modal Reasoning with Context 15 upvotes, #8 of 2025-07-02
  9. Data Efficacy for Language Model Training 10 upvotes, #9 of 2025-07-02
  10. MusiXQA: Advancing Visual Music Understanding in Multimodal Large Language Models 10 upvotes, #9 of 2025-07-02
  11. Training for X-Ray Vision: Amodal Segmentation, Amodal Content Completion, and View-Invariant Object Representation from Multi-Camera Video 10 upvotes, #9 of 2025-07-02
  12. FreeLong++: Training-Free Long Video Generation via Multi-band SpectralFusion 7 upvotes, #12 of 2025-07-02
  13. IR3D-Bench: Evaluating Vision-Language Model Scene Understanding as Agentic Inverse Rendering 5 upvotes, #13 of 2025-07-02
  14. Peccavi: Visual Paraphrase Attack Safe and Distortion Free Image Watermarking Technique for AI-Generated Images 4 upvotes, #14 of 2025-07-02
  15. Confident Splatting: Confidence-Based Compression of 3D Gaussian Splatting via Learnable Beta Distributions 3 upvotes, #15 of 2025-07-02
  16. Mixture of Reasonings: Teach Large Language Models to Reason with Adaptive Strategies 3 upvotes, #15 of 2025-07-02
  17. Ella: Embodied Social Agents with Lifelong Memory 2 upvotes, #17 of 2025-07-02
  18. FreNBRDF: A Frequency-Rectified Neural Material Representation 2 upvotes, #17 of 2025-07-02

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.