一句话：提出从诊断报告直接生成可执行 playbook 的端到端微服务修复任务与基准 MicroRemed，并用经验-仿真强化微调训练 E2E-REME 模型。
微服务系统规模与复杂度持续增长，故障愈发频繁且代价高昂。现有基于 LLM 的自动修复方法主要将文本指令翻译为可执行的 Ansible playbook，依赖专家手工编写的提示词，缺乏运行时知识引导，并需依赖大规模通用 LLM，限制了准确性与效率。论文提出端到端微服务修复（E2E-MR）新任务：直接由诊断报告生成可执行 playbook，自主恢复故障系统；并构建 MicroRemed 基准，自动化完成微服务部署、故障注入、playbook 执行与修复后验证。作者进一步提出 E2E-REME，一种通过经验-仿真强化微调（experience-simulation reinforcement fine-tuning）训练的端到端自动修复模型。在公开与工业微服务平台上的实验，与九个代表性 LLM 相比，E2E-REME 取得了更优的准确性与效率。
Contemporary microservice systems continue to grow in scale and complexity, leading to increasingly frequent and costly failures. While recent LLM-based auto-remediation approaches have emerged, they primarily translate textual instructions into executable Ansible playbooks and rely on expert-crafted prompts, lacking runtime knowledge guidance and depending on large-scale general-purpose LLMs, which limits their accuracy and efficiency. We introduce End-to-End Microservice Remediation (E2E-MR), a new task that requires directly generating executable playbooks from diagnosis reports to autonomously restore faulty systems. To enable rigorous evaluation, we build MicroRemed, a benchmark that automates microservice deployment, failure injection, playbook execution, and post-repair verification. We further propose E2E-REME, an end-to-end auto-remediation model trained via experience-simulation reinforcement fine-tuning. Experiments on public and industrial microservice platforms, compared with nine representative LLMs, show that E2E-REME achieves superior accuracy and efficiency.
DOI 原文 · arXiv · PDF（开放获取） · DBLP
@inproceedings{abs-2604-11094,
title = {E2E-REME: Towards End-to-End Microservices Auto-Remediation via Experience-Simulation Reinforcement Fine-Tuning},
author = {Lingzhe Zhang and Yunpeng Zhai and Tong Jia and Minghua He and Chiming Duan and Zhaoyang Liu and Bolin Ding and Ying Li},
booktitle = {FSE},
year = {2026},
doi = {10.1145/3803437.3805206}
}