策略内增量蒸馏
On-Policy Delta Distillation
浏览论文内容
中文总结 AI 辅助
研究基于策略内蒸馏，引入新的增量信号作为蒸馏奖励，提出策略内增量蒸馏方法。实验表明该方法能显著改进策略内蒸馏，使推理语言模型在短训练期内表现出色，优于传统方法。
中文摘要 AI 辅助
策略内蒸馏是强化学习中的一种训练后方法，通过教师模型提供的token级监督来减轻奖励模型的限制。尽管已在各种场景中研究和应用，但基本设计仍未充分探索。本文引入一种新的蒸馏奖励——增量信号，它是教师模型与其在推理能力指令调整前的基础模型之间的差异。实验表明，增量信号显著改进了策略内蒸馏，新方法在数学、科学和代码推理基准测试中始终优于传统方法，能让推理语言模型在短训练期内实现高性能。
英文摘要
On-policy distillation is an alternative post-training method in reinforcement learning that alleviates the constraints imposed by reward models by providing token-level supervision from a teacher model. Although on-policy distillation has been studied and applied across various settings, its fundamental design remains underexplored. In this paper, we introduce a new distillation reward, termed the delta signal, instead of directly imitating the teacher's output distribution. The delta signal is defined as the difference between the teacher model and its base model prior to instruction tuning for reasoning capability. It therefore captures the changes induced by reasoning tuning and provides a more direct signal for transferring reasoning capabilities. Using extensive empirical evidence, we show that the delta signal substantially improves on-policy distillation and refer to the new distillation method as On-Policy Delta Distillation (OPD$^2$). Experiments across mathematics, science, and code-reasoning benchmarks demonstrate that OPD$^2$ consistently outperforms conventional on-policy distillation, enabling reasoning LLMs to achieve strong performance with only a short post-training period. Code will be available at https://github.com/naver-ai/opd2
发表机构
- NAVER AI Lab（NAVER人工智能实验室）
机构由 AI 辅助整理，请以论文原文为准。
