PR-OPD: Privileged Representation On-policy Self-Distillation for Agentic Reinforcement Learning
Language-model agents are usually trained by reinforcement learning from one reward per episode, and privileged self-distillation enriches it by letting the same policy, given a skill, teach its skill-free self through token probabilities. However, we identify two phenomena that question this channel. Invisible Advanta...
Mu-Yang Li, Jie Yang, Zheng-Yu Fang et al.
· 1 citation
· ⚡1