Daily Papers of 2026-01-26

  1. LongCat-Flash-Thinking-2601 Technical Report 171 upvotes, #1 of 2026-01-26
  2. SWE-Pruner: Self-Adaptive Context Pruning for Coding Agents 87 upvotes, #2 of 2026-01-26
  3. TwinBrainVLA: Unleashing the Potential of Generalist VLMs for Embodied Tasks via Asymmetric Mixture-of-Transformers 60 upvotes, #3 of 2026-01-26
  4. VisGym: Diverse, Customizable, Scalable Environments for Multimodal Agents 40 upvotes, #4 of 2026-01-26
  5. Memory-V2V: Augmenting Video-to-Video Diffusion Models with Memory 28 upvotes, #5 of 2026-01-26
  6. Jet-RL: Enabling On-Policy FP8 Reinforcement Learning with Unified Training and Rollout Precision Flow 21 upvotes, #6 of 2026-01-26
  7. Inference-Time Scaling of Verification: Self-Evolving Deep Research Agents via Test-Time Rubric-Guided Verification 20 upvotes, #7 of 2026-01-26
  8. Endless Terminals: Scaling RL Environments for Terminal Agents 16 upvotes, #8 of 2026-01-26
  9. SALAD: Achieve High-Sparsity Attention via Efficient Linear Attention Tuning for Video Diffusion Transformer 15 upvotes, #9 of 2026-01-26
  10. Dancing in Chains: Strategic Persuasion in Academic Rebuttal via Theory of Mind 13 upvotes, #10 of 2026-01-26
  11. GameTalk: Training LLMs for Strategic Conversation 12 upvotes, #11 of 2026-01-26
  12. MeepleLM: A Virtual Playtester Simulating Diverse Subjective Experiences 11 upvotes, #12 of 2026-01-26
  13. ChartVerse: Scaling Chart Reasoning via Reliable Programmatic Synthesis from Scratch 10 upvotes, #13 of 2026-01-26
  14. DSGym: A Holistic Framework for Evaluating and Training Data Science Agents 10 upvotes, #13 of 2026-01-26
  15. Mecellem Models: Turkish Models Trained from Scratch and Continually Pre-trained for the Legal Domain 7 upvotes, #15 of 2026-01-26
  16. Knowledge is Not Enough: Injecting RL Skills for Continual Adaptation 6 upvotes, #16 of 2026-01-26
  17. VISTA-PATH: An interactive foundation model for pathology image segmentation and quantitative analysis in computational pathology 2 upvotes, #17 of 2026-01-26
  18. Guidelines to Prompt Large Language Models for Code Generation: An Empirical Characterization 1 upvotes, #18 of 2026-01-26

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.