Paper abstract and links
The authors’ abstract, vote count and code link of any paper on the current daily list.
On the current list
- Trajectory2Gym: Decision-Grounded Compositional Interaction Synthesis for Tool-Use Agents
- StreamVibe: Proactive Video Agents for Evolving User Intent
- Trial by Trace: Benchmarking AI Agent Harnesses for Reliability and Failure Attribution
- InteractiveWorld: A Verifiable Human-Centric Environment for Building Better Interactive MLLM Assistants
- WorldCast: World-Consistent Character Video Generation with Large-Angle Camera Control
- Spatial Audio Generation and Editing with Executable Plans and Self-Distilling Transfusion
- DynaFlow: Learning Free-Form Multi-Agent Orchestration
- Work Agents: Systems, Capabilities, and Optimization
- Eat Your Precision: Precision Objective Mitigates Reward Hacking in Rubric-based RL
- World-in-Context: Interactive Video World Modeling with In-Context Conditioning
- Principia: Test-Time Hypotheses to Principle Construction for Scientific Discovery
- Δτ-Bench: Pushing Agents Toward a Moving Capability
- Qwen-Image-PE: Learning Visual Design for Text-to-Image Prompt Enhancement
- Trustworthy Recursive Self-Improvement in the Era of Large Models: Foundation, Risks, Evaluation and Open Challenges
- Continual Learning Exam: Benchmarking Long-Horizon RSI and Experiential Learning in LLM Agents
- RynnVLA-Latent: Scaling Robotic Manipulation Foundation Models with 3D-Grounded Latent Actions
- S3AC: Stabilized, Selective, and Scalable Actor-Critic for LLM Reasoning
- ProAS: A Progress-Aware Advantage Shaping Framework for Multi-Turn Agentic Reinforcement Learning
- Novel or Not? PatEx-Gym: Agentic Prior-Art Search for Novelty Assessment in Patent Examination
- Extreme Compression of Per-Layer Embeddings via Cross-Layer Latent Vector Quantization
- HiP3R: Hierarchical Pose Graphs for Long-Horizon 3D Reconstruction
- AgentProofBench: Benchmarking Reusable Verifiers for Web Agents
- Build, Don't Play: Self-Improving LLM Research Agents Develop a Hearthstone Battlegrounds AI That Rivals Top Players
- Skill^{infty}: Recursive Self-Improvement for Agent Skill Optimization
- miniTE: Diagnosing and Bridging the Execution Gap in Native NVFP4 Reinforcement Learning
- Steering Black-Box Large Language Model Agents with Pitfall Anticipation World Model