Daily Papers of 2025-10-20

  1. A Theoretical Study on Bridging Internal Probability and Self-Consistency for LLM Reasoning 139 upvotes, #1 of 2025-10-20
  2. OmniVinci: Enhancing Architecture and Data for Omni-Modal Understanding LLM 79 upvotes, #2 of 2025-10-20
  3. NANO3D: A Training-Free Approach for Efficient 3D Editing Without Masks 56 upvotes, #3 of 2025-10-20
  4. Scaling Instruction-Based Video Editing with a High-Quality Synthetic Dataset 49 upvotes, #4 of 2025-10-20
  5. Latent Diffusion Model without Variational Autoencoder 47 upvotes, #5 of 2025-10-20
  6. Emergent Misalignment via In-Context Learning: Narrow in-context examples can produce broadly misaligned LLMs 45 upvotes, #6 of 2025-10-20
  7. Skyfall-GS: Synthesizing Immersive 3D Urban Scenes from Satellite Imagery 42 upvotes, #7 of 2025-10-20
  8. Paper2Web: Let's Make Your Paper Alive! 24 upvotes, #8 of 2025-10-20
  9. Language Models Model Language 23 upvotes, #9 of 2025-10-20
  10. LightsOut: Diffusion-based Outpainting for Enhanced Lens Flare Removal 23 upvotes, #9 of 2025-10-20
  11. A^2FM: An Adaptive Agent Foundation Model for Tool-Aware Hybrid Reasoning 22 upvotes, #11 of 2025-10-20
  12. BLIP3o-NEXT: Next Frontier of Native Image Generation 21 upvotes, #12 of 2025-10-20
  13. MorphoBench: A Benchmark with Difficulty Adaptive to Model Reasoning 19 upvotes, #13 of 2025-10-20
  14. VISTA: A Test-Time Self-Improving Video Generation Agent 18 upvotes, #14 of 2025-10-20
  15. DLER: Doing Length pEnalty Right - Incentivizing More Intelligence per Token via Reinforcement Learning 15 upvotes, #15 of 2025-10-20
  16. Foundation Models for Scientific Discovery: From Paradigm Enhancement to Paradigm Transition 14 upvotes, #16 of 2025-10-20
  17. Build Your Personalized Research Group: A Multiagent Framework for Continual and Interactive Science Automation 14 upvotes, #16 of 2025-10-20
  18. Explore to Evolve: Scaling Evolved Aggregation Logic via Proactive Online Exploration for Deep Research Agents 11 upvotes, #18 of 2025-10-20
  19. InfiMed-ORBIT: Aligning LLMs on Open-Ended Complex Tasks via Rubric-Based Incremental Training 10 upvotes, #19 of 2025-10-20
  20. Imaginarium: Vision-guided High-Quality 3D Scene Layout Generation 9 upvotes, #20 of 2025-10-20
  21. FinTrust: A Comprehensive Benchmark of Trustworthiness Evaluation in Finance Domain 5 upvotes, #21 of 2025-10-20
  22. Do LLMs "Feel"? Emotion Circuits Discovery and Control 4 upvotes, #22 of 2025-10-20
  23. Robust Layerwise Scaling Rules by Proper Weight Decay Tuning 4 upvotes, #22 of 2025-10-20
  24. ERGO: Entropy-guided Resetting for Generation Optimization in Multi-turn Language Models 3 upvotes, #24 of 2025-10-20
  25. Rewiring Experts on the Fly:Continuous Rerouting for Better Online Adaptation in Mixture-of-Expert models 3 upvotes, #24 of 2025-10-20
  26. Adapting Self-Supervised Representations as a Latent Space for Efficient Generation 2 upvotes, #26 of 2025-10-20
  27. Train a Unified Multimodal Data Quality Classifier with Synthetic Data 2 upvotes, #26 of 2025-10-20
  28. DriveGen3D: Boosting Feed-Forward Driving Scene Generation with Efficient Video Diffusion 1 upvotes, #28 of 2025-10-20

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.