Xiusi Chen
XtremSup
AI & ML interests
RL, post training
Recent Activity
upvoted a paper 3 days ago
Learning Meta-Skills for Agent Harness Design in Test-Time AI4AI upvoted a paper about 1 month ago
Cliff: Learning Process Rewards from the First Mistake upvoted a paper about 2 months ago
AI4AI at Test-Time: Strong-to-Weak Capability Transfer via HarnessesOrganizations
