CLIFT: Conformal Self-Verification for Web Agent Training and Test-Time Scaling
CLIFT turns judge feedback into reusable rewards and judge-free test-time selection for web agents.
1 stories, newest first · one card per story
CLIFT turns judge feedback into reusable rewards and judge-free test-time selection for web agents.
Today 1.6K readers · 2.5K views · 779 reads · all time 189K views