Daily Papers of 2026-05-22

  1. DelTA: Discriminative Token Credit Assignment for Reinforcement Learning from Verifiable Rewards 204 upvotes, #1 of 2026-05-22
  2. TransitLM: A Large-Scale Dataset and Benchmark for Map-Free Transit Route Generation 174 upvotes, #2 of 2026-05-22
  3. Perception or Prejudice: Can MLLMs Go Beyond First Impressions of Personality? 169 upvotes, #3 of 2026-05-22
  4. π-Bench: Evaluating Proactive Personal Assistant Agents in Long-Horizon Workflows 102 upvotes, #4 of 2026-05-22
  5. Full Attention Strikes Back: Transferring Full Attention into Sparse within Hundred Training Steps 93 upvotes, #5 of 2026-05-22
  6. ACC: Compiling Agent Trajectories for Long-Context Training 59 upvotes, #6 of 2026-05-22
  7. PhysX-Omni: Unified Simulation-Ready Physical 3D Generation for Rigid, Deformable, and Articulated Objects 51 upvotes, #7 of 2026-05-22
  8. LatentOmni: Rethinking Omni-Modal Understanding via Unified Audio-Visual Latent Reasoning 46 upvotes, #8 of 2026-05-22
  9. Forecasting Scientific Progress with Artificial Intelligence 42 upvotes, #9 of 2026-05-22
  10. WorldKV: Efficient World Memory with World Retrieval and Compression 41 upvotes, #10 of 2026-05-22
  11. SEGA: Spectral-Energy Guided Attention for Resolution Extrapolation in Diffusion Transformers 40 upvotes, #11 of 2026-05-22
  12. Spreadsheet-RL: Advancing Large Language Model Agents on Realistic Spreadsheet Tasks via Reinforcement Learning 35 upvotes, #12 of 2026-05-22
  13. Gated DeltaNet-2: Decoupling Erase and Write in Linear Attention 30 upvotes, #13 of 2026-05-22
  14. FlowLong: Inference-time Long Video Generation via Manifold-constrained Tweedie Matching 29 upvotes, #14 of 2026-05-22
  15. SpaceDG: Benchmarking Spatial Intelligence under Visual Degradation 28 upvotes, #15 of 2026-05-22
  16. Sensor2Sensor: Cross-Embodiment Sensor Conversion for Autonomous Driving 27 upvotes, #16 of 2026-05-22
  17. Unsupervised Process Reward Models 25 upvotes, #17 of 2026-05-22
  18. Q-ARVD: Quantizing Autoregressive Video Diffusion Models 21 upvotes, #18 of 2026-05-22
  19. Maestro: Reinforcement Learning to Orchestrate Hierarchical Model-Skill Ensembles 20 upvotes, #19 of 2026-05-22
  20. AutoRubric-T2I: Robust Rule-Based Reward Model for Text-to-Image Alignment 19 upvotes, #20 of 2026-05-22
  21. Training Large Language Models to Predict Clinical Events 17 upvotes, #21 of 2026-05-22
  22. Forecasting Downstream Performance of LLMs With Proxy Metrics 14 upvotes, #22 of 2026-05-22
  23. GenEvolve: Self-Evolving Image Generation Agents via Tool-Orchestrated Visual Experience Distillation 13 upvotes, #23 of 2026-05-22
  24. KVServe: Service-Aware KV Cache Compression for Communication-Efficient Disaggregated LLM Serving 12 upvotes, #24 of 2026-05-22
  25. ClinSeekAgent: Automating Multimodal Evidence Seeking for Agentic Clinical Reasoning 12 upvotes, #24 of 2026-05-22
  26. Bernini: Latent Semantic Planning for Video Diffusion 12 upvotes, #24 of 2026-05-22
  27. Efficient Agentic Reasoning Through Self-Regulated Simulative Planning 11 upvotes, #27 of 2026-05-22
  28. Swift Sampling: Selecting Temporal Surprises via Taylor Series 11 upvotes, #27 of 2026-05-22
  29. One Sentence, One Drama: Personalized Short-Form Drama Generation via Multi-Agent Systems 10 upvotes, #29 of 2026-05-22
  30. LoREnc: Low-Rank Encryption for Securing Foundation Models and LoRA Adapters 9 upvotes, #30 of 2026-05-22
  31. TerminalWorld: Benchmarking Agents on Real-World Terminal Tasks 9 upvotes, #30 of 2026-05-22
  32. Rule2DRC: Benchmarking LLM Agents for DRC Script Synthesis with Execution-Guided Test Generation 8 upvotes, #32 of 2026-05-22
  33. SceneAligner: 3D-Grounded Floorplan Localization in the Wild 8 upvotes, #32 of 2026-05-22
  34. "I didn't Make the Micro Decisions": Measuring, Inducing, and Exposing Goal-Level AI Contributions in Collaboration 6 upvotes, #34 of 2026-05-22
  35. Segment Anything with Motion, Geometry, and Semantic Adaptation for Complex Nonlinear Visual Object Tracking 6 upvotes, #34 of 2026-05-22
  36. Diversed Model Discovery via Structured Table Discovery 6 upvotes, #34 of 2026-05-22
  37. OmniPro: A Comprehensive Benchmark for Omni-Proactive Streaming Video Understanding 5 upvotes, #37 of 2026-05-22
  38. DecQ: Detail-Condensing Queries for Enhanced Reconstruction and Generation in Representation Autoencoders 5 upvotes, #37 of 2026-05-22
  39. SAM 3D Animal: Promptable Animal 3D Reconstruction from Images in the Wild 4 upvotes, #39 of 2026-05-22
  40. Lean Refactor: Multi-Objective Controllable Proof Optimization via Agentic Strategy Search 4 upvotes, #39 of 2026-05-22
  41. Platonic Representations in the Human Brain: Unsupervised Recovery of Universal Geometry 4 upvotes, #39 of 2026-05-22
  42. Same Architecture, Different Capacity: Optimizer-Induced Spectral Scaling Laws 4 upvotes, #39 of 2026-05-22
  43. From Reasoning Chains to Verifiable Subproblems: Curriculum Reinforcement Learning Enables Credit Assignment for LLM Reasoning 4 upvotes, #39 of 2026-05-22
  44. More Context, Larger Models, or Moral Knowledge? A Systematic Study of Schwartz Value Detection in Political Texts 4 upvotes, #39 of 2026-05-22
  45. AnyMo: Geometry-Aware Setup-Agnostic Modeling of Human Motion in the Wild 4 upvotes, #39 of 2026-05-22
  46. Disentangling Sampling from Training Budget in Class-Imbalanced CT Body Composition Segmentation 3 upvotes, #46 of 2026-05-22
  47. FashionLens: Toward Versatile Fashion Image Retrieval via Task-Adaptive Learning 2 upvotes, #47 of 2026-05-22
  48. Live Music Diffusion Models: Efficient Fine-Tuning and Post-Training of Interactive Diffusion Music Generators 2 upvotes, #47 of 2026-05-22
  49. Minimalist Visual Inertial Odometry 1 upvotes, #49 of 2026-05-22

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.