AI.info

Junchi Yan

Explore Junchi Yan on AI.info.

• HandEdit: A Unified Benchmark for Egocentric Human-to-Robot Dexterous Hand Image Editing
• GRADE: Benchmarking Discipline-Informed Reasoning in Image Editing
• ACE-Brain-0: Spatial Intelligence as a Shared Scaffold for Universal Embodiments
• Efficient-LVSM: Faster, Cheaper, and Better Large View Synthesis Model via Decoupled Co-Refinement Attention
• LoPA: Scaling dLLM Inference via Lookahead Parallel Decoding
• Probing Scientific General Intelligence of LLMs with Scientist-Aligned Workflows
• SoliReward: Mitigating Susceptibility to Reward Hacking and Annotation Noise in Video Generation Reward Models
• Dual-Branch Center-Surrounding Contrast: Rethinking Contrastive Learning for 3D Point Clouds
• LaGen: Towards Autoregressive LiDAR Scene Generation
• When AI Agents Collude Online: Financial Fraud Risks by Collaborative LLM Agents on Social Platforms
• NTKMTL: Mitigating Task Imbalance in Multi-Task Learning from Neural Tangent Kernel Perspective
• Attention Illuminates LLM Reasoning: The Preplan-and-Anchor Rhythm Enables Fine-Grained Policy Optimization