Why On-Policy Distillation Sometimes Fails: Vanishing Learning Signals
On-policy distillation (OPD) enables effective capability transfer between language models, yet the mechanisms underlying its failures are not fully understood. Across code generation and mathematical reasoning, OPD with larger-scale teachers exhibits early loss plateaus, with an average final loss reduction of 25.1% a...
Lei Zhao, Qi-Chao Zhao, Bo-Wen Zuo et al.
· 0 citations