Daily Papers of 2025-07-01

  1. Ovis-U1 Technical Report 59 upvotes, #1 of 2025-07-01
  2. SPIRAL: Self-Play on Zero-Sum Games Incentivizes Reasoning via Multi-Agent Multi-Turn Reinforcement Learning 43 upvotes, #2 of 2025-07-01
  3. VMoBA: Mixture-of-Block Attention for Video Diffusion Models 31 upvotes, #3 of 2025-07-01
  4. Calligrapher: Freestyle Text Image Customization 31 upvotes, #3 of 2025-07-01
  5. Listener-Rewarded Thinking in VLMs for Image Preferences 24 upvotes, #5 of 2025-07-01
  6. MEMFOF: High-Resolution Training for Memory-Efficient Multi-Frame Optical Flow Estimation 24 upvotes, #5 of 2025-07-01
  7. Evolving Prompts In-Context: An Open-ended, Self-replicating Perspective 19 upvotes, #7 of 2025-07-01
  8. SparseLoRA: Accelerating LLM Fine-Tuning with Contextual Sparsity 17 upvotes, #8 of 2025-07-01
  9. Consistent Time-of-Flight Depth Denoising via Graph-Informed Geometric Attention 14 upvotes, #9 of 2025-07-01
  10. MARBLE: A Hard Benchmark for Multimodal Spatial Reasoning and Planning 12 upvotes, #10 of 2025-07-01
  11. Aha Moment Revisited: Are VLMs Truly Capable of Self Verification in Inference-time Scaling? 11 upvotes, #11 of 2025-07-01
  12. RExBench: Can coding agents autonomously implement AI research extensions? 11 upvotes, #11 of 2025-07-01
  13. UrbanLLaVA: A Multi-modal Large Language Model for Urban Intelligence with Spatial Reasoning and Understanding 7 upvotes, #13 of 2025-07-01
  14. ThinkSound: Chain-of-Thought Reasoning in Multimodal Large Language Models for Audio Generation and Editing 6 upvotes, #14 of 2025-07-01
  15. RoboScape: Physics-informed Embodied World Model 5 upvotes, #15 of 2025-07-01
  16. Teaching a Language Model to Speak the Language of Tools 4 upvotes, #16 of 2025-07-01
  17. Tower+: Bridging Generality and Translation Specialization in Multilingual LLMs 3 upvotes, #17 of 2025-07-01
  18. VOCABTRIM: Vocabulary Pruning for Efficient Speculative Decoding in LLMs 3 upvotes, #17 of 2025-07-01
  19. Degradation-Modeled Multipath Diffusion for Tunable Metalens Photography 2 upvotes, #19 of 2025-07-01

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.