---
id: 20260731-T0-08
title: "弱到强同策略蒸馏：用弱教师训练更强学生模型"
title_en: "Weak-to-Strong On-Policy Distillation Trains Stronger Students from Weak Teachers"
url: https://ai.daily.yangsir.net/daily/20260731-T0-08
issue_date: 2026-07-31
publish_date: 2026-07-30T04:00:00.000Z
category: research
source_name: "arXiv cs.LG (ML)"
source_url: https://arxiv.org/abs/2607.26246
---

# 弱到强同策略蒸馏：用弱教师训练更强学生模型

arXiv新论文提出“弱到强同策略蒸馏”方法，允许使用能力较弱的教师模型，通过学生自身轨迹进行同策略蒸馏，有效提升学生模型表现。这使得模型能力迁移不再依赖强教师。

## English Version

**Weak-to-Strong On-Policy Distillation Trains Stronger Students from Weak Teachers**

A new paper introduces Weak-to-Strong On-Policy Distillation, which uses weaker teacher models to train stronger student LLMs via student's own rollouts. It enables capability transfer without relying on a strong teacher.

---

**来源**：[arXiv cs.LG (ML)](https://arxiv.org/abs/2607.26246)

**详情页**：https://ai.daily.yangsir.net/daily/20260731-T0-08

---

*智语观潮 · Daily — https://ai.daily.yangsir.net/llms.txt*