Daily Papers of 2025-09-25

  1. Video models are zero-shot learners and reasoners 89 upvotes, #1 of 2025-09-25
  2. SIM-CoT: Supervised Implicit Chain-of-Thought 35 upvotes, #2 of 2025-09-25
  3. EmbeddingGemma: Powerful and Lightweight Text Representations 33 upvotes, #3 of 2025-09-25
  4. EditVerse: Unifying Image and Video Editing and Generation with In-Context Learning 16 upvotes, #4 of 2025-09-25
  5. Advancing Speech Understanding in Speech-Aware Language Models with GRPO 14 upvotes, #5 of 2025-09-25
  6. LLMs4All: A Review on Large Language Models for Research and Applications in Academic Disciplines 10 upvotes, #6 of 2025-09-25
  7. PhysCtrl: Generative Physics for Controllable and Physics-Grounded Video Generation 10 upvotes, #6 of 2025-09-25
  8. Lavida-O: Elastic Large Masked Diffusion Models for Unified Multimodal Understanding and Generation 9 upvotes, #8 of 2025-09-25
  9. SimpleFold: Folding Proteins is Simpler than You Think 7 upvotes, #9 of 2025-09-25
  10. Logics-Parsing Technical Report 5 upvotes, #10 of 2025-09-25
  11. Mixture of Thoughts: Learning to Aggregate What Experts Think, Not Just What They Say 5 upvotes, #10 of 2025-09-25
  12. On the Use of Agentic Coding: An Empirical Study of Pull Requests on GitHub 3 upvotes, #12 of 2025-09-25
  13. ATLAS: Benchmarking and Adapting LLMs for Global Trade via Harmonized Tariff Code Classification 2 upvotes, #13 of 2025-09-25
  14. kh2d-solver: A Python Library for Idealized Two-Dimensional Incompressible Kelvin-Helmholtz Instability 3 upvotes, #14 of 2025-09-25

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.