Daily Papers of 2026-08-07

  1. AgentOPSD: Recursive Self-Distillation for Agentic Reinforcement Learning 94 upvotes, #1 of 2026-08-07
  2. OSReward: Instituting Standardized Evaluation for Cross-Platform Computer-Use Reward Models 72 upvotes, #2 of 2026-08-07
  3. Interpretable MEG Decoding of Perceived Speech: Cortical Sources and the Stimulus Features That Drive Retrieval 69 upvotes, #3 of 2026-08-07
  4. WorldClaw: Agentic 3D Open-World Generation at Scale 68 upvotes, #4 of 2026-08-07
  5. GST-Bench: Can VLMs Develop Global Spatial Awareness from Video? 46 upvotes, #5 of 2026-08-07
  6. EnvACE: Internalizing Environment Dynamics via World Rehearsal for Agentic Reinforcement Learning 43 upvotes, #6 of 2026-08-07
  7. ChronoVision: Temporal Reasoning via Latent State Reconstruction 40 upvotes, #7 of 2026-08-07
  8. Learning from Failures: Retrieval-Centric CoT via Hard Negatives for Unified Multimodal Retrieval 40 upvotes, #7 of 2026-08-07
  9. From Economic Agents to Agentic Economies: A Systems Blueprint for Economic World Models 34 upvotes, #9 of 2026-08-07
  10. HarnessOpt-Bench: Evaluating LLMs at Harness Optimization 34 upvotes, #9 of 2026-08-07
  11. DataSpace: Benchmarking Data Agents for Verifiable Analytics over Heterogeneous Workspaces 33 upvotes, #11 of 2026-08-07
  12. On-Policy Delta Distillation for Multilingual Math Reasoning 32 upvotes, #12 of 2026-08-07
  13. Teaching Nemotron Greek: Mining a Corpus, Adapting Retrieval, and Grounding Generation for Modern Greek across Specialist Domains 31 upvotes, #13 of 2026-08-07
  14. Activity Frames: Deterministic Screen-Activity Compilation for Agent Memory and Replay 28 upvotes, #14 of 2026-08-07
  15. KVAE: Family of Tokenizers for Multimodal Generative Models 28 upvotes, #14 of 2026-08-07
  16. SmartMage: Dynamic Modality Orchestration for 3D Scene Understanding 27 upvotes, #16 of 2026-08-07
  17. World-to-Wrist: Task-Conditioned Future Wrist Modeling for Fine-Grained Robot Manipulation 26 upvotes, #17 of 2026-08-07
  18. PaDoc: Layout-Grounded Parallel Decoding for Document Parsing 24 upvotes, #18 of 2026-08-07
  19. Continual Learning in Transition 23 upvotes, #19 of 2026-08-07
  20. CalibForge: Adversarial Solver Calibration for Scaling Learnable Terminal Tasks 23 upvotes, #19 of 2026-08-07
  21. DyPES-VLA: Learning Shared Dynamics Priors and Embodiment-Specific Control for Cross-Embodiment Manipulation 23 upvotes, #19 of 2026-08-07
  22. EffectLearner: World-Aware Object-Effect Reasoning for Real-World Video Object Removal 22 upvotes, #22 of 2026-08-07
  23. MameLoshnLM: Yiddish Language Model and Evaluation Benchmark 22 upvotes, #22 of 2026-08-07
  24. Invisible Shortcuts: Why Vision Encoders Know Your Camera 17 upvotes, #24 of 2026-08-07
  25. ContextMaster: Interactive Multi-Shot Video Creation via Fixed-Budget Sparse Context Routing 16 upvotes, #25 of 2026-08-07
  26. MASS: Multiplayer World Models with Authoritative Shared State 16 upvotes, #25 of 2026-08-07
  27. FactorJEPA: Factorizing Monolithic Futures into Layout-Agent-Interaction Channels for Crowded and Chaotic Global South Urban Worlds 13 upvotes, #27 of 2026-08-07
  28. GaussianSelector: Lightweight Human-Guided Object Selection in 3D Gaussian Splatting with Graph Optimization 13 upvotes, #27 of 2026-08-07
  29. Weights or Skills? A Survey of Robot-Learning Techniques: from Action-Predicting Weights to Robots that Write their Own Skills 11 upvotes, #29 of 2026-08-07
  30. Task-Conditional Flow Matching for Balanced Multilingual Text Embedding Adaptation 10 upvotes, #30 of 2026-08-07

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.