arxiv:2505.17955
Seong Joon Oh
AI & ML interests
Scalable Trustworthy AI
https://stai-lab.org/
Recent Activity
upvoted a paper 2 days ago
Principia: Test-Time Hypotheses to Principle Construction for Scientific Discovery upvoted a paper 2 days ago
Eat Your Precision: Precision Objective Mitigates Reward Hacking in Rubric-based RL upvoted a paper 6 days ago
Self-Distilling Future Video Representations for World Action Models