Paper recorded by Signals 4 on 2026-08-31 in cs.LG. Abstract reproduced from arXiv; link to the original below.
Published 2026-08-31 on arXiv · recorded by Signals 4 on 2026-09-01
Category: cs.LG · 机器学习 · first seen 2026-09-01
On-policy distillation (OPD) offers dense token-level supervision as an alternative to the sparse outcome-level advantages of reinforcement learning with verifiable rewards (RLVR). However, the teacher scores student-generated trajectories that are inherently off-policy for it, so the reliability of its supervision, and hence the source of the student's improvement, remains unclear. We quantitativ