← 研踪 PaperTrailarXiv:2605.07396 · v2
Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis
首次提交 2026-05-08 15:52
版本更新 2026-11-10 11:40
雷达观测 2026-11-11 11:04
摘要
On-policy distillation (OPD) is a powerful paradigm for LLM post-training, yet its reliance on teacher logits restricts it to white-box teachers. In this paper, we study whether rubric-based supervision can provide a text-only feedback interface for black-box OPD, and what makes it effective. Our instantiation, Rubric-based On-Policy Distillation (ROPD), induces rubrics by contrasting teacher responses with the student's current rollouts and uses them to score these rollouts for on-policy optimization. We find that, under matched settings, text-only ROPD outperforms logit-based OPD methods on four competition-math benchmarks, reaching the best AIME24 accuracy of standard logit-based OPD with 9.6x fewer training samples. Our analysis yields three findings: rubric rewards track correctness far better than teacher log-likelihood (AUC 0.90 vs. 0.35); training with them repairs more failed criteria and regresses fewer passed ones than logit-based supervision; and contrasting teacher responses with student rollouts is central to rubric quality. These results position rubric-based OPD as a flexible, black-box-compatible alternative to logit-based OPD. Our implementation is available at https://github.com/Peregrine123/ROPD_official.
作者备注：Preprint. Code is available at https://github.com/Peregrine123/ROPD_official
观测到的版本
相邻记录可能跳过中间版本；首次发现 v2 及以上也会标为版本更新。
v2 · 2026-11-10 11:40 · 最新观测版本
Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis
On-policy distillation (OPD) is a powerful paradigm for LLM post-training, yet its reliance on teacher logits restricts it to white-box teachers. In this paper, we study whether rubric-based supervision can provide a text-only feedback interface for black-box OPD, and what makes it effective. Our instantiation, Rubric-based On-Policy Distillation (ROPD), induces rubrics by contrasting teacher responses with the student's current rollouts and uses them to score these rollouts for on-policy optimization. We find that, under matched settings, text-only ROPD outperforms logit-based OPD methods on four competition-math benchmarks, reaching the best AIME24 accuracy of standard logit-based OPD with 9.6x fewer training samples. Our analysis yields three findings: rubric rewards track correctness far better than teacher log-likelihood (AUC 0.90 vs. 0.35); training with them repairs more failed criteria and regresses fewer passed ones than logit-based supervision; and contrasting teacher responses with student rollouts is central to rubric quality. These results position rubric-based OPD as a flexible, black-box-compatible alternative to logit-based OPD. Our implementation is available at https://github.com/Peregrine123/ROPD_official.
查看此版本解读 →查看此版本原文 ↗