Daily Papers of 2026-04-21

  1. Extending One-Step Image Generation from Class Labels to Text via Discriminative Text Representation 96 upvotes, #1 of 2026-04-21
  2. OneVL: One-Step Latent Reasoning and Planning with Vision-Language Explanation 87 upvotes, #2 of 2026-04-21
  3. Agent-World: Scaling Real-World Environment Synthesis for Evolving General Agent Intelligence 80 upvotes, #3 of 2026-04-21
  4. OpenGame: Open Agentic Coding for Games 74 upvotes, #4 of 2026-04-21
  5. MultiWorld: Scalable Multi-Agent Multi-View Video World Models 43 upvotes, #5 of 2026-04-21
  6. EasyVideoR1: Easier RL for Video Understanding 40 upvotes, #6 of 2026-04-21
  7. ClawEnvKit: Automatic Environment Generation for Claw-Like Agents 28 upvotes, #7 of 2026-04-21
  8. When Can LLMs Learn to Reason with Weak Supervision? 24 upvotes, #8 of 2026-04-21
  9. GFT: From Imitation to Reward Fine-Tuning with Unbiased Group Advantages and Dynamic Coefficient Rectification 23 upvotes, #9 of 2026-04-21
  10. SkillFlow:Benchmarking Lifelong Skill Discovery and Evolution for Autonomous Agents 22 upvotes, #10 of 2026-04-21
  11. WebCompass: Towards Multimodal Web Coding Evaluation for Code Language Models 22 upvotes, #10 of 2026-04-21
  12. Crowded in B-Space: Calibrating Shared Directions for LoRA Merging 18 upvotes, #12 of 2026-04-21
  13. The Illusion of Certainty: Decoupling Capability and Calibration in On-Policy Distillation 14 upvotes, #13 of 2026-04-21
  14. MathNet: a Global Multimodal Benchmark for Mathematical Reasoning and Retrieval 14 upvotes, #13 of 2026-04-21
  15. Concrete Jungle: Towards Concreteness Paved Contrastive Negative Mining for Compositional Understanding 12 upvotes, #15 of 2026-04-21
  16. GenericAgent: A Token-Efficient Self-Evolving LLM Agent via Contextual Information Density Maximization (V1.0) 12 upvotes, #15 of 2026-04-21
  17. On the Reliability of Computer Use Agents 11 upvotes, #17 of 2026-04-21
  18. Meta-learning In-Context Enables Training-Free Cross Subject Brain Decoding 9 upvotes, #18 of 2026-04-21
  19. Training LLM Agents for Spontaneous, Reward-Free Self-Evolution via World Knowledge Exploration 9 upvotes, #18 of 2026-04-21
  20. VoxMind: An End-to-End Agentic Spoken Dialogue System 8 upvotes, #20 of 2026-04-21
  21. OmniScript: Towards Audio-Visual Script Generation for Long-Form Cinematic Video 7 upvotes, #21 of 2026-04-21
  22. Agents Explore but Agents Ignore: LLMs Lack Environmental Curiosity 7 upvotes, #21 of 2026-04-21
  23. Revisiting a Pain in the Neck: A Semantic Reasoning Benchmark for Language Models 6 upvotes, #23 of 2026-04-21
  24. Beyond Text-Dominance: Understanding Modality Preference of Omni-modal Large Language Models 6 upvotes, #23 of 2026-04-21
  25. Stratagem: Learning Transferable Reasoning via Trajectory-Modulated Game Self-Play 6 upvotes, #23 of 2026-04-21
  26. Multiplication in Multimodal LLMs: Computation with Text, Image, and Audio Inputs 6 upvotes, #23 of 2026-04-21
  27. River-LLM: Large Language Model Seamless Exit Based on KV Share 6 upvotes, #23 of 2026-04-21
  28. Precise Debugging Benchmark: Is Your Model Debugging or Regenerating? 4 upvotes, #28 of 2026-04-21
  29. EvoMaster: A Foundational Agent Framework for Building Evolving Autonomous Scientific Agents at Scale 4 upvotes, #28 of 2026-04-21
  30. MARCO: Navigating the Unseen Space of Semantic Correspondence 4 upvotes, #28 of 2026-04-21
  31. MedConclusion: A Benchmark for Biomedical Conclusion Generation from Structured Abstracts 3 upvotes, #31 of 2026-04-21
  32. Forge-UGC: FX optimization and register-graph engine for universal graph compiler 3 upvotes, #31 of 2026-04-21
  33. Geometric coherence of single-cell CRISPR perturbations reveals regulatory architecture and predicts cellular stress 3 upvotes, #31 of 2026-04-21
  34. When Background Matters: Breaking Medical Vision Language Models by Transferable Attack 3 upvotes, #31 of 2026-04-21
  35. MTR-DuplexBench: Towards a Comprehensive Evaluation of Multi-Round Conversations for Full-Duplex Speech Language Models 2 upvotes, #35 of 2026-04-21
  36. Protecting Language Models Against Unauthorized Distillation through Trace Rewriting 2 upvotes, #35 of 2026-04-21
  37. Symbolic Guardrails for Domain-Specific Agents: Stronger Safety and Security Guarantees Without Sacrificing Utility 2 upvotes, #35 of 2026-04-21
  38. Modeling Sparse and Bursty Vulnerability Sightings: Forecasting Under Data Constraints 2 upvotes, #35 of 2026-04-21
  39. MNAFT: modality neuron-aware fine-tuning of multimodal large language models for image translation 2 upvotes, #35 of 2026-04-21
  40. The Continuity Layer: Why Intelligence Needs an Architecture for What It Carries Forward 2 upvotes, #35 of 2026-04-21
  41. Back to Repair: A Minimal Denoising Network\ for Time Series Anomaly Detection 2 upvotes, #35 of 2026-04-21
  42. The Geometric Canary: Predicting Steerability and Detecting Drift via Representational Stability 2 upvotes, #35 of 2026-04-21
  43. Latent Preference Modeling for Cross-Session Personalized Tool Calling 2 upvotes, #35 of 2026-04-21
  44. Modeling Multiple Support Strategies within a Single Turn for Emotional Support Conversations 2 upvotes, #35 of 2026-04-21
  45. Significance and Stability Analysis of Gene-Environment Interaction using RGxEStat 1 upvotes, #45 of 2026-04-21
  46. KWBench: Measuring Unprompted Problem Recognition in Knowledge Work 1 upvotes, #45 of 2026-04-21
  47. On the Robustness of LLM-Based Dense Retrievers: A Systematic Analysis of Generalizability and Stability 1 upvotes, #45 of 2026-04-21
  48. Terminal Wrench: A Dataset of 331 Reward-Hackable Environments and 3,632 Exploit Trajectories 1 upvotes, #45 of 2026-04-21
  49. HSG: Hyperbolic Scene Graph 1 upvotes, #49 of 2026-04-21

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.