Beyond Teacher Likelihood: Group-Calibrated On-Policy Distillation for Long-Context Reasoning

By Zhu Zhang · Paper · cs.LG

On-policy distillation (OPD) trains a student on its own responses using dense token-level guidance from a stronger teacher. In long-context tasks, however, token-level teacher support can favor locally plausible responses that omit evidence distributed across the input or violat

Cs.lg

View original

HomeResourceLoading…