Daily Papers of 2025-11-25

  1. General Agentic Memory Via Deep Research 150 upvotes, #1 of 2025-11-25
  2. AutoEnv: Automated Environments for Measuring Cross-Environment Agent Learning 88 upvotes, #2 of 2025-11-25
  3. DeCo: Frequency-Decoupled Pixel Diffusion for End-to-End Image Generation 62 upvotes, #3 of 2025-11-25
  4. DR Tulu: Reinforcement Learning with Evolving Rubrics for Deep Research 53 upvotes, #4 of 2025-11-25
  5. Computer-Use Agents as Judges for Generative User Interface 50 upvotes, #5 of 2025-11-25
  6. UltraFlux: Data-Model Co-Design for High-quality Native 4K Text-to-Image Generation across Diverse Aspect Ratios 37 upvotes, #6 of 2025-11-25
  7. In-Video Instructions: Visual Signals as Generative Control 28 upvotes, #7 of 2025-11-25
  8. The Image as Its Own Reward: Reinforcement Learning with Adversarial Reward for Image Generation 26 upvotes, #8 of 2025-11-25
  9. Chain-of-Visual-Thought: Teaching VLMs to See and Think Better with Continuous Visual Tokens 25 upvotes, #9 of 2025-11-25
  10. Budget-Aware Tool-Use Enables Effective Agent Scaling 24 upvotes, #10 of 2025-11-25
  11. HunyuanVideo 1.5 Technical Report 21 upvotes, #11 of 2025-11-25
  12. Pillar-0: A New Frontier for Radiology Foundation Models 19 upvotes, #12 of 2025-11-25
  13. Multi-Agent Deep Research: Training Multi-Agent Systems with M-GRPO 17 upvotes, #13 of 2025-11-25
  14. M3-Bench: Multi-Modal, Multi-Hop, Multi-Threaded Tool-Using MLLM Agent Benchmark 16 upvotes, #14 of 2025-11-25
  15. Plan-X: Instruct Video Generation via Semantic Planning 16 upvotes, #14 of 2025-11-25
  16. Beyond Multiple Choice: Verifiable OpenQA for Robust Vision-Language RFT 10 upvotes, #16 of 2025-11-25
  17. One4D: Unified 4D Generation and Reconstruction via Decoupled LoRA Control 10 upvotes, #16 of 2025-11-25
  18. Controllable Layer Decomposition for Reversible Multi-Layer Image Generation 8 upvotes, #18 of 2025-11-25
  19. MIST: Mutual Information Via Supervised Training 8 upvotes, #18 of 2025-11-25
  20. AICC: Parse HTML Finer, Make Models Better -- A 7.3T AI-Ready Corpus Built by a Model-Based HTML Parser 7 upvotes, #20 of 2025-11-25
  21. Upsample Anything: A Simple and Hard to Beat Baseline for Feature Upsampling 6 upvotes, #21 of 2025-11-25
  22. PRInTS: Reward Modeling for Long-Horizon Information Seeking 6 upvotes, #21 of 2025-11-25
  23. MASS: Motion-Aware Spatial-Temporal Grounding for Physics Reasoning and Comprehension in Vision-Language Models 5 upvotes, #23 of 2025-11-25
  24. EvoVLA: Self-Evolving Vision-Language-Action Model 4 upvotes, #24 of 2025-11-25
  25. Flow Map Distillation Without Data 4 upvotes, #24 of 2025-11-25
  26. Target-Bench: Can World Models Achieve Mapless Path Planning with Semantic Targets? 3 upvotes, #26 of 2025-11-25
  27. Extracting Interaction-Aware Monosemantic Concepts in Recommender Systems 1 upvotes, #27 of 2025-11-25
  28. Fidelity-Aware Recommendation Explanations via Stochastic Path Integration 1 upvotes, #27 of 2025-11-25
  29. Representational Stability of Truth in Large Language Models 1 upvotes, #27 of 2025-11-25
  30. SyncMV4D: Synchronized Multi-view Joint Diffusion of Appearance and Motion for Hand-Object Interaction Synthesis 1 upvotes, #27 of 2025-11-25
  31. MSRNet: A Multi-Scale Recursive Network for Camouflaged Object Detection 1 upvotes, #31 of 2025-11-25

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.