Daily Papers of 2026-08-03

  1. From RLVR to RLSVR: Task Transformation Induces Self-Verifiable Rewards for Open-Ended LLM Self-Improvement 104 upvotes, #1 of 2026-08-03
  2. Mental World Modeling 103 upvotes, #2 of 2026-08-03
  3. N_0-VTLA: Scaling Vision-Tactile-Language-Action Model with Latent Tactile Tokens 76 upvotes, #3 of 2026-08-03
  4. Weak-to-Strong On-Policy Distillation 56 upvotes, #4 of 2026-08-03
  5. Meshy T2: Fast Native Mesh Generation with Flow Matching 56 upvotes, #4 of 2026-08-03
  6. N_0-TWAM: Scaling Tactile-Native World-Action Model for Contact-Rich Manipulation 47 upvotes, #6 of 2026-08-03
  7. Scaling Properties of Text Conditioning in Visual Generation 38 upvotes, #7 of 2026-08-03
  8. AISPA: User-Centric System Prompt Auditing for Large Language Model Applications 37 upvotes, #8 of 2026-08-03
  9. SAF-OPD: Stable Advantage Fusion for On-Policy Distillation 34 upvotes, #9 of 2026-08-03
  10. Fewer Clarifications, Better Code: Benchmarking Cross-Session Personalized Ambiguity Adaptation in Coding Assistants 32 upvotes, #10 of 2026-08-03
  11. QQWorld: Quantile-Quantile Matching for World Model Regularization 30 upvotes, #11 of 2026-08-03
  12. Enhancing Rubric-based RL via Self-Distillation 25 upvotes, #12 of 2026-08-03
  13. ExtractBench: A Benchmark for Schema-Guided Enterprise Document Extraction 23 upvotes, #13 of 2026-08-03
  14. Evaluation-Verification Reward for Consistent Multi-Reference Image Editing 16 upvotes, #14 of 2026-08-03
  15. ODEWorld: A Continuous Predictive Architecture via Physical-Time Flow 14 upvotes, #15 of 2026-08-03
  16. EMBL AI Librarian: Life-Sciences Knowledge Layer for AI Agents 12 upvotes, #16 of 2026-08-03
  17. RL^2-VLA: Adaptive RL Latent Compositional Steering with Test-Time Scaling for Vision-Language-Action Models 9 upvotes, #17 of 2026-08-03
  18. One Future, Every Robot: Label-Efficient Collective-State Prediction with Decentralized JEPA 9 upvotes, #17 of 2026-08-03
  19. Not All Tokens Deserve Equal Credit: Counterfactual Sensitivity Credit Reallocation for Long-CoT Reasoning 8 upvotes, #19 of 2026-08-03
  20. In the Driver's Seat: A Multi-Company Study on the Reality of Autonomous Driving System Testing 7 upvotes, #20 of 2026-08-03
  21. Constitutional Midtraining: Content Presence Drives Alignment Gains 7 upvotes, #20 of 2026-08-03
  22. Safeguards Based on Copyable Context Cannot Provide Reliable Safety for LLMs 7 upvotes, #20 of 2026-08-03
  23. SULAND v2: A Refined RGB Dataset and Deep Learning Object Detection Benchmark for UAV/UGV-Based SUrface LANDmine Detection Under Domain Shift 6 upvotes, #23 of 2026-08-03
  24. Would You Walk to the Car Wash? Revealing the Salience Bias of Large Language Models in Commonsense Reasoning 5 upvotes, #24 of 2026-08-03
  25. Toward Robust and 3D-Aware RGB-NIR Imaging in the Dark 5 upvotes, #24 of 2026-08-03
  26. Beyond Feeling Better: Capability-Sustaining Emotional Dialogue as a Longitudinal Research Paradigm 4 upvotes, #26 of 2026-08-03
  27. SGTP: Sampling-based Game-Theoretic Planning for Real-Time Multi-Vehicle Autonomous Racing 1 upvotes, #27 of 2026-08-03

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.