Daily Papers of 2025-10-07

  1. Paper2Video: Automatic Video Generation from Scientific Papers 99 upvotes, #1 of 2025-10-07
  2. Agentic Context Engineering: Evolving Contexts for Self-Improving Language Models 89 upvotes, #2 of 2025-10-07
  3. Video-LMM Post-Training: A Deep Dive into Video Reasoning with Large Multimodal Models 43 upvotes, #3 of 2025-10-07
  4. MITS: Enhanced Tree Search Reasoning for LLMs via Pointwise Mutual Information 38 upvotes, #4 of 2025-10-07
  5. VChain: Chain-of-Visual-Thought for Reasoning in Video Generation 34 upvotes, #5 of 2025-10-07
  6. Imperceptible Jailbreaking against Large Language Models 33 upvotes, #6 of 2025-10-07
  7. Hybrid Architectures for Language Models: Systematic Analysis and Design Insights 31 upvotes, #7 of 2025-10-07
  8. Optimal Scaling Needs Optimal Norm 28 upvotes, #8 of 2025-10-07
  9. Reactive Transformer (RxT) -- Stateful Real-Time Processing for Event-Driven Reactive Language Models 23 upvotes, #9 of 2025-10-07
  10. Front-Loading Reasoning: The Synergy between Pretraining and Post-Training Data 21 upvotes, #10 of 2025-10-07
  11. MOSS-Speech: Towards True Speech-to-Speech Models Without Text Guidance 18 upvotes, #11 of 2025-10-07
  12. Fine-Tuning on Noisy Instructions: Effects on Generalization and Performance 15 upvotes, #12 of 2025-10-07
  13. Factuality Matters: When Image Generation and Editing Meet Structured Visuals 15 upvotes, #12 of 2025-10-07
  14. Reinforce-Ada: An Adaptive Sampling Framework for Reinforce-Style LLM Training 14 upvotes, #14 of 2025-10-07
  15. Judging with Confidence: Calibrating Autoraters to Preference Distributions 13 upvotes, #15 of 2025-10-07
  16. Prosperity before Collapse: How Far Can Off-Policy RL Reach with Stale Data on LLMs? 12 upvotes, #16 of 2025-10-07
  17. SwiReasoning: Switch-Thinking in Latent and Explicit for Pareto-Superior Reasoning LLMs 11 upvotes, #17 of 2025-10-07
  18. Self-Reflective Generation at Test Time 9 upvotes, #18 of 2025-10-07
  19. ChronoEdit: Towards Temporal Reasoning for Image Editing and World Simulation 9 upvotes, #18 of 2025-10-07
  20. Watch and Learn: Learning to Use Computers from Online Videos 9 upvotes, #18 of 2025-10-07
  21. Code4MeV2: a Research-oriented Code-completion Platform 7 upvotes, #21 of 2025-10-07
  22. EvolProver: Advancing Automated Theorem Proving by Evolving Formalized Problems via Symmetry and Difficulty 5 upvotes, #22 of 2025-10-07
  23. Good Intentions Beyond ACL: Who Does NLP for Social Good, and Where? 5 upvotes, #22 of 2025-10-07
  24. Character Mixing for Video Generation 5 upvotes, #22 of 2025-10-07
  25. Optimized Minimal 4D Gaussian Splatting 4 upvotes, #25 of 2025-10-07
  26. Test-Time Scaling in Diffusion LLMs via Hidden Semi-Autoregressive Experts 4 upvotes, #25 of 2025-10-07
  27. HiKE: Hierarchical Evaluation Framework for Korean-English Code-Switching Speech Recognition 3 upvotes, #27 of 2025-10-07
  28. LLMSQL: Upgrading WikiSQL for the LLM Era of Text-to-SQL 3 upvotes, #27 of 2025-10-07
  29. Thai Semantic End-of-Turn Detection for Real-Time Voice Agents 3 upvotes, #27 of 2025-10-07
  30. Slow-Fast Policy Optimization: Reposition-Before-Update for LLM Reasoning 3 upvotes, #27 of 2025-10-07
  31. MoME: Mixture of Matryoshka Experts for Audio-Visual Speech Recognition 3 upvotes, #27 of 2025-10-07
  32. SAEdit: Token-level control for continuous image editing via Sparse AutoEncoder 3 upvotes, #27 of 2025-10-07
  33. Multilingual Routing in Mixture-of-Experts 2 upvotes, #33 of 2025-10-07
  34. Alignment Tipping Process: How Self-Evolution Pushes LLM Agents Off the Rails 2 upvotes, #33 of 2025-10-07
  35. Graph2Eval: Automatic Multimodal Task Generation for Agents via Knowledge Graphs 1 upvotes, #35 of 2025-10-07
  36. AdvEvo-MARL: Shaping Internalized Safety through Adversarial Co-Evolution in Multi-Agent Reinforcement Learning 1 upvotes, #35 of 2025-10-07
  37. Position: Privacy Is Not Just Memorization! 1 upvotes, #35 of 2025-10-07
  38. CWM: An Open-Weights LLM for Research on Code Generation with World Models 1 upvotes, #35 of 2025-10-07
  39. Paris: A Decentralized Trained Open-Weight Diffusion Model 1 upvotes, #35 of 2025-10-07
  40. Epistemic Diversity and Knowledge Collapse in Large Language Models 1 upvotes, #35 of 2025-10-07
  41. Utility-Learning Tension in Self-Modifying Agents 1 upvotes, #35 of 2025-10-07
  42. Learning on the Job: Test-Time Curricula for Targeted Reinforcement Learning 1 upvotes, #35 of 2025-10-07
  43. Federated Computation of ROC and PR Curves 1 upvotes, #43 of 2025-10-07
  44. Power Transform Revisited: Numerically Stable, and Federated 1 upvotes, #43 of 2025-10-07

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.