On-policy distillation

Concept

A technique to mitigate the need for token-level credit assignment by having a student model generate completions, then asking a stronger teacher model to evaluate the reasonableness of each token generated.

Mentioned in 1 video

Videos Mentioning On-policy distillation

Stanford CME295 Transformers & LLMs | Autumn 2026 | Lecture 4 - Reinforcement Learning with LLMs

Stanford CME295 Transformers & LLMs | Autumn 2026 | Lecture 4 - Reinforcement Learning with LLMs

Stanford Online

A technique to mitigate the need for token-level credit assignment by having a student model generate completions, then asking a stronger teacher model to evaluate the reasonableness of each token generated.