Daily Papers of 2025-09-09

  1. Reverse-Engineered Reasoning for Open-Ended Generation 142 upvotes, #1 of 2025-09-09
  2. WebExplorer: Explore and Evolve for Training Long-Horizon Web Agents 76 upvotes, #2 of 2025-09-09
  3. Revolutionizing Reinforcement Learning Framework for Diffusion Large Language Models 53 upvotes, #3 of 2025-09-09
  4. Does DINOv3 Set a New Medical Vision Standard? 34 upvotes, #4 of 2025-09-09
  5. Reinforcement Learning Foundations for Deep Research Systems: A Survey 31 upvotes, #5 of 2025-09-09
  6. Reinforced Visual Perception with Tools 29 upvotes, #6 of 2025-09-09
  7. DivMerge: A divergence-based model merging method for multi-tasking 24 upvotes, #7 of 2025-09-09
  8. Paper2Agent: Reimagining Research Papers As Interactive and Reliable AI Agents 24 upvotes, #7 of 2025-09-09
  9. Focusing by Contrastive Attention: Enhancing VLMs' Visual Reasoning 16 upvotes, #9 of 2025-09-09
  10. UniVerse-1: Unified Audio-Video Generation via Stitching of Experts 13 upvotes, #10 of 2025-09-09
  11. SFR-DeepResearch: Towards Effective Reinforcement Learning for Autonomously Reasoning Single Agents 13 upvotes, #10 of 2025-09-09
  12. Interleaving Reasoning for Better Text-to-Image Generation 13 upvotes, #10 of 2025-09-09
  13. Easier Painting Than Thinking: Can Text-to-Image Models Set the Stage, but Not Direct the Play? 10 upvotes, #13 of 2025-09-09
  14. Scaling up Multi-Turn Off-Policy RL and Multi-Agent Tree Search for LLM Step-Provers 10 upvotes, #13 of 2025-09-09
  15. Guided Decoding and Its Critical Role in Retrieval-Augmented Generation 7 upvotes, #15 of 2025-09-09
  16. Test-Time Scaling in Reasoning Models Is Not Effective for Knowledge-Intensive Tasks Yet 7 upvotes, #15 of 2025-09-09
  17. Inpaint4Drag: Repurposing Inpainting Models for Drag-Based Image Editing via Bidirectional Warping 5 upvotes, #17 of 2025-09-09
  18. Llama-GENBA-10B: A Trilingual Large Language Model for German, English and Bavarian 5 upvotes, #17 of 2025-09-09
  19. D-HUMOR: Dark Humor Understanding via Multimodal Open-ended Reasoning 5 upvotes, #17 of 2025-09-09
  20. R^textbf{2AI}: Towards Resistant and Resilient AI in an Evolving World 3 upvotes, #20 of 2025-09-09
  21. Mechanistic interpretability for steering vision-language-action models 2 upvotes, #21 of 2025-09-09
  22. Singular Value Few-shot Adaptation of Vision-Language Models 2 upvotes, #21 of 2025-09-09
  23. MAS-Bench: A Unified Benchmark for Shortcut-Augmented Hybrid Mobile GUI Agents 2 upvotes, #21 of 2025-09-09
  24. DCReg: Decoupled Characterization for Efficient Degenerate LiDAR Registration 1 upvotes, #24 of 2025-09-09
  25. Saturation-Driven Dataset Generation for LLM Mathematical Reasoning in the TPTP Ecosystem 1 upvotes, #24 of 2025-09-09

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.