Daily Papers of 2025-04-10

  1. DDT: Decoupled Diffusion Transformer 69 upvotes, #1 of 2025-04-10
  2. OLMoTrace: Tracing Language Model Outputs Back to Trillions of Training Tokens 69 upvotes, #1 of 2025-04-10
  3. Missing Premise exacerbates Overthinking: Are Reasoning Models losing Critical Thinking Skill? 35 upvotes, #3 of 2025-04-10
  4. FantasyTalking: Realistic Talking Portrait Generation via Coherent Motion Synthesis 30 upvotes, #4 of 2025-04-10
  5. A Unified Agentic Framework for Evaluating Conditional Image Generation 28 upvotes, #5 of 2025-04-10
  6. GenDoP: Auto-regressive Camera Trajectory Generation as a Director of Photography 21 upvotes, #6 of 2025-04-10
  7. A Sober Look at Progress in Language Model Reasoning: Pitfalls and Paths to Reproducibility 18 upvotes, #7 of 2025-04-10
  8. OmniCaptioner: One Captioner to Rule Them All 17 upvotes, #8 of 2025-04-10
  9. Self-Steering Language Models 15 upvotes, #9 of 2025-04-10
  10. Caption Anything in Video: Fine-grained Object-centric Captioning via Spatiotemporal Multimodal Prompting 14 upvotes, #10 of 2025-04-10
  11. WildGS-SLAM: Monocular Gaussian Splatting SLAM in Dynamic Environments 9 upvotes, #11 of 2025-04-10
  12. VideoChat-R1: Enhancing Spatio-Temporal Perception via Reinforcement Fine-Tuning 9 upvotes, #11 of 2025-04-10
  13. DiTaiListener: Controllable High Fidelity Listener Video Generation with Diffusion 8 upvotes, #13 of 2025-04-10
  14. Masked Scene Modeling: Narrowing the Gap Between Supervised and Self-Supervised Learning in 3D Scene Understanding 8 upvotes, #13 of 2025-04-10
  15. Are We Done with Object-Centric Learning? 5 upvotes, #15 of 2025-04-10
  16. Pretraining Language Models for Diachronic Linguistic Change Discovery 4 upvotes, #16 of 2025-04-10
  17. RobustDexGrasp: Robust Dexterous Grasping of General Objects from Single-view Perception 3 upvotes, #17 of 2025-04-10
  18. RuOpinionNE-2024: Extraction of Opinion Tuples from Russian News Texts 3 upvotes, #17 of 2025-04-10
  19. Fast Controlled Generation from Language Models with Adaptive Weighted Rejection Sampling 2 upvotes, #19 of 2025-04-10

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.