Reinforcing What the Student Has Learned: Teacher-Guided On-Policy Distillation
On-policy distillation (OPD) improves language-model reasoning through dense teacher feedback on student-generated responses.We first observe an amplitude disparity in this feedback: on math reasoning responses, style-related positions exhibit 3.5 times the mean absolute teacher--student log-probability difference of m...
Bo-Zhen Fang, Lin-Hao Zhong, Xue-Hao Wang et al.
· 0 citations