跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.21020 · v1

SWE-OPD: On-Policy Distillation for Agentic Coding

首次提交 2026-11-16 20:39
版本更新 2026-11-16 20:39
雷达观测 2026-11-17 17:11
arXiv ↗PDF ↗

摘要

In this work, we study on-policy distillation (OPD) for transferring agentic coding capabilities from a large teacher model to a smaller student model. Prior work has focused primarily on reasoning and short-horizon agentic tasks. We present SWE-OPD, a minimal OPD study for software engineering, where the student interacts with executable repositories and the teacher supervises the prefixes visited by the student. Across three teacher--student pairs, two training sources, and three SWE benchmarks, we find that on-policy supervision transfers coding capability: supervision from Qwen3.6-27B increases the SWE-bench Pro resolve rate of Qwen3.5-9B from 35.9% to 39.4%. Refreshing trajectories is not strictly required either, as an offline variant with cached teacher scores still improves over the student baselines while cutting end-to-end GPU-hours by 59.4%--72.6% and halving peak GPU allocation, at benchmark-dependent accuracy. Furthermore, we extend the offline variant to corss-tokenizer scenarios: distilling GLM-4.5-Air from DeepSeek-V4-Flash-0731 raises its SWE-bench Pro and Multilingual resolve rate from 35.8% to 39.7% and from 41.8% to 46.2%, around 4% and 4.6% improvement, respectively. We also find that a decreasing loss, however, does not necessarily indicate improved agent performance: under top-KK reverse-KL loss, one student exhibits severe repetition despite a decreasing loss. We identify a support-mass gradient term as a potential mechanism underlying this failure.

作者备注:under review

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

目前只收录了一个版本。收录下一版后,可以对照标题、作者、分类和摘要。

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。

v1 · 2026-11-16 20:39 · 最新观测版本

SWE-OPD: On-Policy Distillation for Agentic Coding

In this work, we study on-policy distillation (OPD) for transferring agentic coding capabilities from a large teacher model to a smaller student model. Prior work has focused primarily on reasoning and short-horizon agentic tasks. We present SWE-OPD, a minimal OPD study for software engineering, where the student interacts with executable repositories and the teacher supervises the prefixes visited by the student. Across three teacher--student pairs, two training sources, and three SWE benchmarks, we find that on-policy supervision transfers coding capability: supervision from Qwen3.6-27B increases the SWE-bench Pro resolve rate of Qwen3.5-9B from 35.9% to 39.4%. Refreshing trajectories is not strictly required either, as an offline variant with cached teacher scores still improves over the student baselines while cutting end-to-end GPU-hours by 59.4%--72.6% and halving peak GPU allocation, at benchmark-dependent accuracy. Furthermore, we extend the offline variant to corss-tokenizer scenarios: distilling GLM-4.5-Air from DeepSeek-V4-Flash-0731 raises its SWE-bench Pro and Multilingual resolve rate from 35.8% to 39.7% and from 41.8% to 46.2%, around 4% and 4.6% improvement, respectively. We also find that a decreasing loss, however, does not necessarily indicate improved agent performance: under top-KK reverse-KL loss, one student exhibits severe repetition despite a decreasing loss. We identify a support-mass gradient term as a potential mechanism underlying this failure.

查看此版本解读 →查看此版本原文 ↗