arxiv:2609.31847
👋 Open to Work
Yanlin Li
yanlinli
AI & ML interests
Trustworthy Agentic AI, Unified Multimodal Models, Human-Centered AI
Recent Activity
upvoted a paper 5 days ago
RSI-Anything: Unlocking Recursive Self-Improvement with World Models upvoted a paper 8 days ago
Omni-IO Skills: Harnessing Your Agent Omni-Native updated a dataset 8 days ago
yanlinli/ours-onlygeo-abl-3dgs-web-assets