arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2512.24040cs.AI
收藏

ROAD: 通过自动调试实现零样本智能体对齐的反思优化

ROAD: Reflective Optimization via Automated Debugging for Zero-Shot Agent Alignment

Amity AI Research and Application Center(Amity人工智能研究与应用中心);Tollring

Natchaya Temyingyong, Daman Jain, Neeraj Kumarsahu, Prabhat Kumar, Rachata Phondi, Wachiravit Modecrua, Krittanon Kaewtawee, Krittin Pachtrachai, Touchapon Kraisingkorn

AI总结:

ROAD通过自动调试实现零样本智能体对齐的反思优化,利用多智能体架构将无结构故障日志转化为结构化决策树协议,提升智能体性能和样本效率。

摘要:

自动提示优化(APO)已成为提升大语言模型(LLM)性能的关键技术,但当前最先进的方法通常依赖于大规模、标注良好的黄金标准开发集来计算进化或强化学习(RL)方法的适应度分数。然而,在现实软件工程中,此类精心编纂的数据集在智能体开发的初始冷启动阶段几乎不可用,工程师只能面对杂乱的生产日志和不断演变的故障模式。我们提出了ROAD(通过自动调试实现的反思优化),一种新颖的框架,通过将优化视为动态调试调查而非随机搜索来绕过对精细数据集的依赖。与传统的突变策略不同,ROAD利用专门的多智能体架构,包括用于根本原因分析的分析器、用于模式聚合的优化器以及用于策略整合的教练,将无结构的故障日志转化为稳健的结构化决策树协议。我们评估了ROAD在标准化学术基准和实时生产知识管理系统上的表现。实验结果表明,ROAD在样本效率方面非常突出,在仅三次自动化迭代中就实现了成功率的5.6%提升(73.6%至79.2%)和搜索准确率的3.8%提升。此外,在零售领域的复杂推理任务中,ROAD相对于基线的智能体性能提高了约19%。这些发现表明,模仿人类工程循环中的故障分析和修补过程为部署可靠LLM智能体提供了一种可行的数据高效替代方案,而不是资源密集型RL训练。

英文摘要:

Automatic Prompt Optimization (APO) has emerged as a critical technique for enhancing Large Language Model (LLM) performance, yet current state-of-the-art methods typically rely on large, labeled gold-standard development sets to compute fitness scores for evolutionary or Reinforcement Learning (RL) approaches. In real-world software engineering, however, such curated datasets are rarely available during the initial cold start of agent development, where engineers instead face messy production logs and evolving failure modes. We present ROAD (Reflective Optimization via Automated Debugging), a novel framework that bypasses the need for refined datasets by treating optimization as a dynamic debugging investigation rather than a stochastic search. Unlike traditional mutation strategies, ROAD utilizes a specialized multi-agent architecture, comprising an Analyzer for root-cause analysis, an Optimizer for pattern aggregation, and a Coach for strategy integration, to convert unstructured failure logs into robust, structured Decision Tree Protocols. We evaluated ROAD across both a standardized academic benchmark and a live production Knowledge Management engine. Experimental results demonstrate that ROAD is highly sample-efficient, achieving a 5.6 percent increase in success rate (73.6 percent to 79.2 percent) and a 3.8 percent increase in search accuracy within just three automated iterations. Furthermore, on complex reasoning tasks in the retail domain, ROAD improved agent performance by approximately 19 percent relative to the baseline. These findings suggest that mimicking the human engineering loop of failure analysis and patching offers a viable, data-efficient alternative to resource-intensive RL training for deploying reliable LLM agents.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑