Another Blueprint In The Wall: How to Ask Frontier AI Like a Kid?

ReactVAU: A Slow-Fast Decoupled Framework for Streaming Video Anomaly Understanding

Stanford Researchers Release Paper2Agent: Turning Research Papers Into AI Agents That Reproduce Results and Run on New Data

Generalized Agent Iteration: One Formal Framework for Iterative Policy Improvement and Recursive Self-Improvement

Marigold V2: Revisiting Diffusion Transformers for Monocular Depth Estimation

AgentGrad: Intervention-guided Prompt Optimization for Multi Agent Systems

ActReview: Rebuttal-Guided Training Data and Rubric Rewards for Actionable Peer Review Generation

SAEScientist-Bench: Can AI Agents Conduct Autonomous SAE Interpretability Research?

Learning 3D Editing without Paired Supervision via Generative Prior Distillation

Knowing What Not to Answer: Selective Non-Compliance in Vision-Language Models

Refuse without Refusal: A Structural Analysis of Safety-Tuning Responses for Reducing False Refusals in Language Models

What Else Needs Fixing? Exploring Cost-Effective Test-Time Compute for Revision Propagation in Artifacts Generated Through Conversation

Scaling Automatic Research Agents via World Models

Raft算法