Daily Papers of 2026-02-13

  1. The Devil Behind Moltbook: Anthropic Safety is Always Vanishing in Self-Evolving AI Societies 188 upvotes, #1 of 2026-02-13
  2. Composition-RL: Compose Your Verifiable Prompts for Reinforcement Learning of Large Language Models 92 upvotes, #2 of 2026-02-13
  3. DeepGen 1.0: A Lightweight Unified Multimodal Model for Advancing Image Generation and Editing 78 upvotes, #3 of 2026-02-13
  4. Learning beyond Teacher: Generalized On-Policy Distillation with Reward Extrapolation 57 upvotes, #4 of 2026-02-13
  5. GigaBrain-0.5M*: a VLA That Learns From World Model-Based Reinforcement Learning 55 upvotes, #5 of 2026-02-13
  6. MOSS-Audio-Tokenizer: Scaling Audio Tokenizers for Future Audio Foundation Models 49 upvotes, #6 of 2026-02-13
  7. NarraScore: Bridging Visual Narrative and Musical Dynamics via Hierarchical Affective Control 43 upvotes, #7 of 2026-02-13
  8. LawThinker: A Deep Research Legal Agent in Dynamic Environments 34 upvotes, #8 of 2026-02-13
  9. Thinking with Drafting: Optical Decompression via Logical Reconstruction 32 upvotes, #9 of 2026-02-13
  10. Stroke of Surprise: Progressive Semantic Illusions in Vector Sketching 31 upvotes, #10 of 2026-02-13
  11. Think Longer to Explore Deeper: Learn to Explore In-Context via Length-Incentivized Reinforcement Learning 30 upvotes, #11 of 2026-02-13
  12. RISE: Self-Improving Robot Policy with Compositional World Model 28 upvotes, #12 of 2026-02-13
  13. χ_{0}: Resource-Aware Robust Manipulation via Taming Distributional Inconsistencies 25 upvotes, #13 of 2026-02-13
  14. EgoHumanoid: Unlocking In-the-Wild Loco-Manipulation with Robot-Free Egocentric Demonstration 20 upvotes, #14 of 2026-02-13
  15. dVoting: Fast Voting for dLLMs 20 upvotes, #14 of 2026-02-13
  16. Sparse Video Generation Propels Real-World Beyond-the-View Vision-Language Navigation 18 upvotes, #16 of 2026-02-13
  17. Voxtral Realtime 15 upvotes, #17 of 2026-02-13
  18. DeepSight: An All-in-One LM Safety Toolkit 13 upvotes, #18 of 2026-02-13
  19. Unveiling Implicit Advantage Symmetry: Why GRPO Struggles with Exploration and Difficulty Adaptation 12 upvotes, #19 of 2026-02-13
  20. Adapting Vision-Language Models for E-commerce Understanding at Scale 12 upvotes, #19 of 2026-02-13
  21. Gaia2: Benchmarking LLM Agents on Dynamic and Asynchronous Environments 12 upvotes, #19 of 2026-02-13
  22. PISCO: Precise Video Instance Insertion with Sparse Control 11 upvotes, #22 of 2026-02-13
  23. T3D: Few-Step Diffusion Language Models via Trajectory Self-Distillation with Direct Discriminative Optimization 8 upvotes, #23 of 2026-02-13
  24. MemFly: On-the-Fly Memory Optimization via Information Bottleneck 7 upvotes, #24 of 2026-02-13
  25. ThinkRouter: Efficient Reasoning via Routing Thinking between Latent and Discrete Spaces 7 upvotes, #24 of 2026-02-13
  26. Single-minus gluon tree amplitudes are nonzero 7 upvotes, #24 of 2026-02-13
  27. Dreaming in Code for Curriculum Learning in Open-Ended Worlds 6 upvotes, #27 of 2026-02-13
  28. MiniCPM-SALA: Hybridizing Sparse and Linear Attention for Efficient Long-Context Modeling 6 upvotes, #27 of 2026-02-13
  29. MolmoSpaces: A Large-Scale Open Ecosystem for Robot Navigation and Manipulation 5 upvotes, #29 of 2026-02-13
  30. MetaphorStar: Image Metaphor Understanding and Reasoning with End-to-End Visual Reinforcement Learning 4 upvotes, #30 of 2026-02-13
  31. Multimodal Fact-Level Attribution for Verifiable Reasoning 4 upvotes, #30 of 2026-02-13
  32. Pretraining A Large Language Model using Distributed GPUs: A Memory-Efficient Decentralized Paradigm 4 upvotes, #30 of 2026-02-13
  33. P-GenRM: Personalized Generative Reward Model with Test-time User-based Scaling 4 upvotes, #30 of 2026-02-13
  34. Sci-CoE: Co-evolving Scientific Reasoning LLMs via Geometric Consensus with Sparse Supervision 4 upvotes, #30 of 2026-02-13
  35. Budget-Constrained Agentic Large Language Models: Intention-Based Planning for Costly Tool Use 3 upvotes, #35 of 2026-02-13
  36. ExStrucTiny: A Benchmark for Schema-Variable Structured Information Extraction from Document Images 3 upvotes, #35 of 2026-02-13
  37. Stemphonic: All-at-once Flexible Multi-stem Music Generation 2 upvotes, #37 of 2026-02-13
  38. Neural Additive Experts: Context-Gated Experts for Controllable Model Additivity 2 upvotes, #37 of 2026-02-13
  39. ABot-N0: Technical Report on the VLA Foundation Model for Versatile Embodied Navigation 2 upvotes, #37 of 2026-02-13
  40. ScalSelect: Scalable Training-Free Multimodal Data Selection for Efficient Visual Instruction Tuning 2 upvotes, #37 of 2026-02-13
  41. Detecting RLVR Training Data via Structural Convergence of Reasoning 2 upvotes, #37 of 2026-02-13

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.