Daily Papers of 2026-06-29

  1. Formalizing Latent Thoughts: Four Axioms of Thought Representation in LLMs 59 upvotes, #1 of 2026-06-29
  2. PhysisForcing: Physics Reinforced World Simulator for Robotic Manipulation 50 upvotes, #2 of 2026-06-29
  3. Qwen-Image-2.0-RL Technical Report 48 upvotes, #3 of 2026-06-29
  4. Translation as a Bridging Action: Transferring Manipulation Skills from Humans to Robots 39 upvotes, #4 of 2026-06-29
  5. Qwen-RobotManip Technical Report: Alignment Unlocks Scale for Robotic Manipulation Foundation Models 28 upvotes, #5 of 2026-06-29
  6. Qwen-RobotNav Technical Report: A Scalable Navigation Model Designed for an Agentic Navigation System 26 upvotes, #6 of 2026-06-29
  7. MultiHashFormer: Hash-based Generative Language Models 21 upvotes, #7 of 2026-06-29
  8. SingGuard: A Policy-Adaptive Multimodal LLM Guardrail with Dynamic Reasoning 15 upvotes, #8 of 2026-06-29
  9. SimFoundry: Modular and Automated Scene Generation for Policy Learning and Evaluation 14 upvotes, #9 of 2026-06-29
  10. GBC: Gradient-Based Connections for Optimizing Multi-Agent Systems 13 upvotes, #10 of 2026-06-29
  11. Thinking While Speaking: Inference-Time Knowledge Transfer for Responsive and Intelligent Conversational Voice Agents 12 upvotes, #11 of 2026-06-29
  12. ProMSA:Progressive Multimodal Search Agents for Knowledge-Based Visual Question Answering 12 upvotes, #11 of 2026-06-29
  13. Vesta: A Generalist Embodied Reasoning Model 10 upvotes, #13 of 2026-06-29
  14. The Tatoxa System for Text Detoxification in Low-Resource Languages: The Case of Tatar 10 upvotes, #13 of 2026-06-29
  15. AgentOdyssey: Open-Ended Long-Horizon Text Game Generation for Test-Time Continual Learning Agents 8 upvotes, #15 of 2026-06-29
  16. Ko-WideSearch: A Korean Breadth-Search Benchmark for Exhaustive Set Enumeration by Web Agents 8 upvotes, #15 of 2026-06-29
  17. Object-Centric Residual RL for Zero-Shot Sim-to-Real VLA Enhancement 7 upvotes, #17 of 2026-06-29
  18. Towards Automating Scientific Review with Google's Paper Assistant Tool 7 upvotes, #17 of 2026-06-29
  19. MemoBench: Benchmarking World Modeling in Dynamically Changing Environments 6 upvotes, #19 of 2026-06-29
  20. Parallel Rollout Approximation for Pixel-Space Autoregressive Image Generation 6 upvotes, #19 of 2026-06-29
  21. Simplified Sparse Attention via Gist Tokens 5 upvotes, #21 of 2026-06-29
  22. Learning to Fold: prizewinning solution at LeHome Challenge 2026 (1st place online, 2nd offline) 5 upvotes, #21 of 2026-06-29
  23. CogniRoute: Learning to Route Social Evidence in Omni-Modal Models 4 upvotes, #23 of 2026-06-29
  24. To Run or Not to Run: Analyzing the Cost-Effectiveness of Code Execution in LLM-Based Program Repair 4 upvotes, #23 of 2026-06-29
  25. Cluster, Route, Escalate: Cascaded Framework for Cost-Aware LLM Serving 4 upvotes, #23 of 2026-06-29
  26. NormGuard: Reward-Preserving Norm Constraints in Flow-Matching Reinforcement Learning 4 upvotes, #23 of 2026-06-29
  27. The Galaxy's Guide to the Tokenizer: A Benchmark for Scientific Foundation Models 3 upvotes, #27 of 2026-06-29
  28. How Much Static Structure Do Code Agents Need? A Study of Deterministic Anchoring 3 upvotes, #27 of 2026-06-29
  29. Boundary-Aware Context Grounding for A Low-Channel EEG Agent 2 upvotes, #29 of 2026-06-29

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.