AI Dynamics

Global AI News Aggregator

About

PUST Uses Smaller Proxy Models for Post-Training Exploration

“Proxy Exploration and Reusable Guidance” Post-training usually forces the large model to do expensive RL exploration itself. This paper, PUST, moves that exploration to a smaller proxy model, then transfers only the learned update direction, not the proxy’s final distribution.

→ View original post on X — @askalphaxiv