# 面向软件工程智能体的类别感知迭代专家训练框架

- 来源：HuggingFace Daily Papers（社区热门论文）
- 发布时间：2026-09-20 08:00
- AIHOT 分数：43
- AIHOT 链接：https://aihot.news/items/cmuc1ig6905eprotsitty325l
- 原文链接：https://arxiv.org/abs/2609.23377

## AI 摘要

针对仓库级软件工程任务中池化强化学习出现的类别此消彼长问题，研究者提出类别感知的专家训练与策略整合框架，通过同源类别专家交替进行长程 Agentic-miniRL 与 Refresh-Repair-Expand（RRE），并用标签路由多教师同策略蒸馏（MOPD）将专家整合为单一可部署学生模型。

## 正文

Repository-level software engineering (SWE) comprises heterogeneous task categories, whose progress under pooled agentic reinforcement learning can be uneven: gains in some categories coincide with regressions in others, while aggregate resolution obscures these changes. Motivated by this category see-saw, we develop a category-aware expert-training and policy-integration framework. Executable task construction and SWE Labeler, an evidence-grounded multi-axis labeling system, organize the training pools. Initial category-specific RL improves average training success while leaving uneven instance-level progress, motivating explicit consolidation of successful behavior and policy-adaptive task selection.

Same-origin category experts alternate long-horizon Agentic-miniRL with Refresh-Repair-Expand (RRE): the updated policy refreshes instance mastery, reuses its own verified successful trajectories for Repair SFT, and reselects tasks for further RL. Label-routed multi-teacher on-policy distillation (MOPD) consolidates the experts into one deployable student, with ReLU-gated reward extrapolation keeping only each teacher's improving direction over the reference. Expert training and policy integration require no external model to provide solution trajectories or action targets.

We evaluate Pooled RL and Balanced RL, expert development, and single-model integration through aggregate and per-category resolution, the minimum category lift over each joint-RL baseline, and expert-gain recovery. The final MOPD policy achieves mean resolution of 58.04% on Pro-618 and 59.00% on SWE-bench Multilingual, improving over the base model by 5.39 and 2.78 percentage points, respectively.