Daily Papers of 2026-07-20

  1. RAGU: A Multi-Step GraphRAG Engine with a Compact Domain-Adapted LLM 146 upvotes, #1 of 2026-07-20
  2. RESOURCE2SKILL: Distilling Executable Agent Skills from Human-Created Multimodal Resources 141 upvotes, #2 of 2026-07-20
  3. Loop the Loopies! 73 upvotes, #3 of 2026-07-20
  4. Xiaomi-Robotics-1: Scaling Vision-Language-Action Models with over 100K Hours of Real-World Trajectories 70 upvotes, #4 of 2026-07-20
  5. xHC: Expanded Hyper-Connections 56 upvotes, #5 of 2026-07-20
  6. Cura 1T: Specialized Model for Agentic Healthcare 52 upvotes, #6 of 2026-07-20
  7. On-Policy Delta Distillation 37 upvotes, #7 of 2026-07-20
  8. RecGPT-V3 Technical Report 30 upvotes, #8 of 2026-07-20
  9. From Human-Centric to Agentic Code Review: The Impact of Different Generations of Generative AI Technology on Review Quality 29 upvotes, #9 of 2026-07-20
  10. Understanding Reasoning from Pretraining to Post-Training 28 upvotes, #10 of 2026-07-20
  11. Qwen-Music Technical Report 27 upvotes, #11 of 2026-07-20
  12. REBASE: Reference-Background Subspace Elimination for Training-Free In-Context Segmentation 26 upvotes, #12 of 2026-07-20
  13. Recursive Harness Self-Improvement 24 upvotes, #13 of 2026-07-20
  14. S1-Omni: A Unified Multimodal Reasoning Model for Scientific Understanding, Prediction, and Generation 16 upvotes, #14 of 2026-07-20
  15. When Does Muon Help Agentic Reinforcement Learning? 15 upvotes, #15 of 2026-07-20
  16. VideoRAE: Taming Video Foundation Models for Generative Modeling via Representation Autoencoders 14 upvotes, #16 of 2026-07-20
  17. Beyond Entropy: Correctness-Aware Advantage Shaping via Contrastive Policy Optimization 12 upvotes, #17 of 2026-07-20
  18. DSWorld: A Data Science World Model for Efficient Autonomous Agents 12 upvotes, #17 of 2026-07-20
  19. Audio-Visual Flamingo: Open Audio-Visual Intelligence for Long and Complex Videos 11 upvotes, #19 of 2026-07-20
  20. Beyond Success Rate: Cost-Aware Evaluation of Offensive and Defensive Security Agents 8 upvotes, #20 of 2026-07-20
  21. See like a Robot: Robot-Centric Pointmaps for Vision-Language-Action Models 7 upvotes, #21 of 2026-07-20
  22. Behavioral Privacy Leakage in Agentic Negotiation: Formalizing and Mitigating Inference Attacks via Randomized Policies 5 upvotes, #22 of 2026-07-20
  23. Agon: Competitive Cross-Model RL with Implicit Rival Grading of Reasoning 5 upvotes, #22 of 2026-07-20
  24. SVR-R1: Bootstrapping Multi-modal Reasoning with Self-verification in Reinforcement Learning 5 upvotes, #22 of 2026-07-20
  25. Benchmarking Sensor Robustness in Plasma Diagnostic Models: A Systematic Evaluation on TokaMark 2 upvotes, #25 of 2026-07-20
  26. Partially Correlated Verifier Cascades in LLM Harnesses: Concave Log-Odds, Polynomial Reliability, and Blind-Spot Ceilings 2 upvotes, #25 of 2026-07-20

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.