MEGA Hub

REOPD: Reliability-Adaptive Reward Extrapolation for On-Policy Distillation

Authors

Do you know Yang Sun?You can claim authorship or link another user.Do you know Lichao Ma?You can claim authorship or link another user.Do you know Houyuan Qin?You can claim authorship or link another user.Do you know Yuxin Liu?You can claim authorship or link another user.Do you know Hanyang Lu?You can claim authorship or link another user.Do you know Yao Zhu?You can claim authorship or link another user.Do you know Pinlong Cai?You can claim authorship or link another user.Do you know Guohang Yan?You can claim authorship or link another user.

Abstract

On-policy distillation (OPD) trains a student on its own trajectories under dense token-level supervision from a teacher. Reward-extrapolation methods such as ExOPD amplify the teacher-reference log-likelihood ratio to move beyond direct imitation, but apply a single global coefficient $λ$ to every token. This can drive the student to fit extreme peaks in the implicit reward, causing reward hacking and unstable training, and the optimal $λ$ varies across domains, requiring costly sweeps. We propose REOPD, a reliability-adaptive reward extrapolation framework for OPD. REOPD combines a token-level compatibility weight with a batch-level adaptive budget, yielding a token-wise coefficient $λ_{b,t}=1+γ_b q_t$ that preserves teacher alignment while selectively extrapolating along reliable teacher-reference directions. It requires no verifier, reward model, value model, or extra rollout beyond standard OPD. REOPD outperforms G-OPD on single-teacher mathematics and on both domains in the multi-teacher setting, while matching G-OPD on single-teacher code, demonstrating effective fine-grained reliability adaptation across domains and teacher configurations.

Community

00