Daily Papers of 2026-02-11

  1. OPUS: Towards Efficient and Principled Data Selection in Large Language Model Pre-training in Every Iteration 315 upvotes, #1 of 2026-02-11
  2. Code2World: A GUI World Model via Renderable Code Generation 189 upvotes, #2 of 2026-02-11
  3. UI-Venus-1.5 Technical Report 149 upvotes, #3 of 2026-02-11
  4. Chain of Mindset: Reasoning with Adaptive Cognitive Modes 70 upvotes, #4 of 2026-02-11
  5. SkillRL: Evolving Agents via Recursive Skill-Augmented Reinforcement Learning 65 upvotes, #5 of 2026-02-11
  6. P1-VL: Bridging Visual Perception and Scientific Reasoning in Physics Olympiads 57 upvotes, #6 of 2026-02-11
  7. Agent World Model: Infinity Synthetic Environments for Agentic Reinforcement Learning 48 upvotes, #7 of 2026-02-11
  8. Prism: Spectral-Aware Block-Sparse Attention 35 upvotes, #8 of 2026-02-11
  9. DLLM-Searcher: Adapting Diffusion Large Language Model for Search Agents 30 upvotes, #9 of 2026-02-11
  10. Agent Banana: High-Fidelity Image Editing with Agentic Thinking and Tooling 27 upvotes, #10 of 2026-02-11
  11. Olaf-World: Orienting Latent Actions for Video World Modeling 26 upvotes, #11 of 2026-02-11
  12. Dr. MAS: Stable Reinforcement Learning for Multi-Agent LLM Systems 24 upvotes, #12 of 2026-02-11
  13. TokenTrim: Inference-Time Token Pruning for Autoregressive Long Video Generation 21 upvotes, #13 of 2026-02-11
  14. Secure Code Generation via Online Reinforcement Learning with Vulnerability Reward Model 20 upvotes, #14 of 2026-02-11
  15. SCALE: Self-uncertainty Conditioned Adaptive Looking and Execution for Vision-Language-Action Models 19 upvotes, #15 of 2026-02-11
  16. Condition Errors Refinement in Autoregressive Image Generation with Diffusion Loss 19 upvotes, #15 of 2026-02-11
  17. LatentLens: Revealing Highly Interpretable Visual Tokens in LLMs 18 upvotes, #17 of 2026-02-11
  18. VLA-JEPA: Enhancing Vision-Language-Action Model with Latent World Model 17 upvotes, #18 of 2026-02-11
  19. BagelVLA: Enhancing Long-Horizon Manipulation via Interleaved Vision-Language-Action Generation 16 upvotes, #19 of 2026-02-11
  20. Large-Scale Terminal Agentic Trajectory Generation from Dockerized Environments 15 upvotes, #20 of 2026-02-11
  21. iGRPO: Self-Feedback-Driven LLM Reasoning 15 upvotes, #20 of 2026-02-11
  22. VideoWorld 2: Learning Transferable Knowledge from Real-world Videos 14 upvotes, #22 of 2026-02-11
  23. ScaleEnv: Scaling Environment Synthesis from Scratch for Generalist Interactive Tool-Use Agent Training 13 upvotes, #23 of 2026-02-11
  24. Fine-T2I: An Open, Large-Scale, and Diverse Dataset for High-Quality T2I Fine-Tuning 13 upvotes, #23 of 2026-02-11
  25. Contact-Anchored Policies: Contact Conditioning Creates Strong Robot Utility Models 12 upvotes, #25 of 2026-02-11
  26. Effective Reasoning Chains Reduce Intrinsic Dimensionality 11 upvotes, #26 of 2026-02-11
  27. Steer2Adapt: Dynamically Composing Steering Vectors Elicits Efficient Adaptation of LLMs 10 upvotes, #27 of 2026-02-11
  28. Dynamic Long Context Reasoning over Compressed Memory via End-to-End Reinforcement Learning 10 upvotes, #27 of 2026-02-11
  29. MIND: Benchmarking Memory Consistency and Action Control in World Models 9 upvotes, #29 of 2026-02-11
  30. Rethinking Global Text Conditioning in Diffusion Transformers 8 upvotes, #30 of 2026-02-11
  31. Covo-Audio Technical Report 8 upvotes, #30 of 2026-02-11
  32. SAGE: Scalable Agentic 3D Scene Generation for Embodied AI 7 upvotes, #32 of 2026-02-11
  33. TodoEvolve: Learning to Architect Agent Planning Systems 6 upvotes, #33 of 2026-02-11
  34. TreeCUA: Efficiently Scaling GUI Automation with Tree-Structured Verifiable Evolution 6 upvotes, #33 of 2026-02-11
  35. ANCHOR: Branch-Point Data Generation for GUI Agents 5 upvotes, #35 of 2026-02-11
  36. Learning to Continually Learn via Meta-learning Agentic Memory Designs 5 upvotes, #35 of 2026-02-11
  37. OPE: Overcoming Information Saturation in Parallel Thinking via Outline-Guided Path Exploration 5 upvotes, #35 of 2026-02-11
  38. Autoregressive Image Generation with Masked Bit Modeling 5 upvotes, #35 of 2026-02-11
  39. SceneSmith: Agentic Generation of Simulation-Ready Indoor Scenes 5 upvotes, #35 of 2026-02-11
  40. Locas: Your Models are Principled Initializers of Locally-Supported Parametric Memories 4 upvotes, #40 of 2026-02-11
  41. From Directions to Regions: Decomposing Activations in Language Models via Local Geometry 3 upvotes, #41 of 2026-02-11
  42. Stable Velocity: A Variance Perspective on Flow Matching 3 upvotes, #41 of 2026-02-11
  43. ContextBench: A Benchmark for Context Retrieval in Coding Agents 3 upvotes, #41 of 2026-02-11
  44. Stop the Flip-Flop: Context-Preserving Verification for Fast Revocable Diffusion Decoding 3 upvotes, #41 of 2026-02-11
  45. Learning Self-Correction in Vision-Language Models via Rollout Augmentation 3 upvotes, #41 of 2026-02-11
  46. On the Optimal Reasoning Length for RL-Trained Language Models 3 upvotes, #41 of 2026-02-11
  47. AgentSys: Secure and Dynamic LLM Agents Through Explicit Hierarchical Memory Management 2 upvotes, #47 of 2026-02-11
  48. Learning on the Manifold: Unlocking Standard Diffusion Transformers with Representation Encoders 2 upvotes, #47 of 2026-02-11
  49. SHARP: Social Harm Analysis via Risk Profiles for Measuring Inequities in Large Language Models 1 upvotes, #49 of 2026-02-11
  50. SafePred: A Predictive Guardrail for Computer-Using Agents via World Models 1 upvotes, #49 of 2026-02-11
  51. C-ΔΘ: Circuit-Restricted Weight Arithmetic for Selective Refusal 1 upvotes, #49 of 2026-02-11
  52. VISTA-Bench: Do Vision-Language Models Really Understand Visualized Text as Well as Pure Text? 1 upvotes, #49 of 2026-02-11
  53. Temporal Pair Consistency for Variance-Reduced Flow Matching 1 upvotes, #49 of 2026-02-11
  54. Surprisal-Guided Selection: Compute-Optimal Test-Time Strategies for Execution-Grounded Code Generation 1 upvotes, #49 of 2026-02-11
  55. CausalArmor: Efficient Indirect Prompt Injection Guardrails via Causal Attribution 1 upvotes, #49 of 2026-02-11
  56. Bridging Academia and Industry: A Comprehensive Benchmark for Attributed Graph Clustering 1 upvotes, #49 of 2026-02-11
  57. LLMs Encode Their Failures: Predicting Success from Pre-Generation Activations 1 upvotes, #49 of 2026-02-11

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.