Daily Papers of 2026-04-09

  1. Think in Strokes, Not Pixels: Process-Driven Image Generation via Interleaved Reasoning 70 upvotes, #1 of 2026-04-09
  2. RAGEN-2: Reasoning Collapse in Agentic RL 63 upvotes, #2 of 2026-04-09
  3. MARS: Enabling Autoregressive Models Multi-Token Generation 38 upvotes, #3 of 2026-04-09
  4. INSPATIO-WORLD: A Real-Time 4D World Simulator via Spatiotemporal Autoregressive Modeling 35 upvotes, #4 of 2026-04-09
  5. FP4 Explore, BF16 Train: Diffusion Reinforcement Learning via Efficient Rollout Scaling 34 upvotes, #5 of 2026-04-09
  6. SEVerA: Verified Synthesis of Self-Evolving Agents 31 upvotes, #6 of 2026-04-09
  7. Combee: Scaling Prompt Learning for Self-Improving Language Model Agents 30 upvotes, #7 of 2026-04-09
  8. Neural Computers 29 upvotes, #8 of 2026-04-09
  9. Qualixar OS: A Universal Operating System for AI Agent Orchestration 16 upvotes, #9 of 2026-04-09
  10. TC-AE: Unlocking Token Capacity for Deep Compression Autoencoders 16 upvotes, #9 of 2026-04-09
  11. Graph-Based Chain-of-Thought Pruning for Reducing Redundant Reflections in Reasoning LLMs 13 upvotes, #11 of 2026-04-09
  12. Personalized RewardBench: Evaluating Reward Models with Human Aligned Personalization 13 upvotes, #11 of 2026-04-09
  13. Beyond Hard Negatives: The Importance of Score Distribution in Knowledge Distillation for Dense Retrieval 12 upvotes, #13 of 2026-04-09
  14. The Depth Ceiling: On the Limits of Large Language Models in Discovering Latent Planning 11 upvotes, #14 of 2026-04-09
  15. A Frame is Worth One Token: Efficient Generative World Modeling with Delta Tokens 10 upvotes, #15 of 2026-04-09
  16. AgentGL: Towards Agentic Graph Learning with LLMs via Reinforcement Learning 10 upvotes, #15 of 2026-04-09
  17. FlowInOne:Unifying Multimodal Generation as Image-in, Image-out Flow Matching 10 upvotes, #15 of 2026-04-09
  18. Learning to Hint for Reinforcement Learning 9 upvotes, #18 of 2026-04-09
  19. DeonticBench: A Benchmark for Reasoning over Rules 9 upvotes, #18 of 2026-04-09
  20. Improving Semantic Proximity in Information Retrieval through Cross-Lingual Alignment 9 upvotes, #18 of 2026-04-09
  21. Q-Zoom: Query-Aware Adaptive Perception for Efficient Multimodal Large Language Models 8 upvotes, #21 of 2026-04-09
  22. R3PM-Net: Real-time, Robust, Real-world Point Matching Network 7 upvotes, #22 of 2026-04-09
  23. MoRight: Motion Control Done Right 7 upvotes, #22 of 2026-04-09
  24. Fast Spatial Memory with Elastic Test-Time Training 7 upvotes, #22 of 2026-04-09
  25. On the Step Length Confounding in LLM Reasoning Data Selection 6 upvotes, #25 of 2026-04-09
  26. Tunable Soft Equivariance with Guarantees 5 upvotes, #26 of 2026-04-09
  27. A Systematic Study of Cross-Modal Typographic Attacks on Audio-Visual Reasoning 4 upvotes, #27 of 2026-04-09
  28. VenusBench-Mobile: A Challenging and User-Centric Benchmark for Mobile GUI Agents with Capability Diagnostics 4 upvotes, #27 of 2026-04-09
  29. GenLCA: 3D Diffusion for Full-Body Avatars from In-the-Wild Videos 4 upvotes, #27 of 2026-04-09

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.