用置信预测方法精准定位多智能体系统错误源头，支持自动回滚修复。
Conformal Agent Error Attribution
- 基于置信预测构建连续序列的错误区间，适合追踪智能体交互轨迹。
- 在多个数据集上验证可精确隔离错误位置，实现高效系统回滚。
- 无需依赖具体模型，适用于各类大语言模型驱动的多智能体系统。
当多智能体系统（MAS）发生故障时，识别导致错误的关键位置是实现自动化恢复至先前状态的第一步。由于基于大语言模型的多智能体系统生成的交互轨迹过长，错误归因仍是根本挑战。本文提出一种基于置信预测（Conformal Prediction, CP）的错误归因框架，提供有限样本、分布无关的覆盖率保证。我们设计了适用于序列数据（如智能体轨迹）的新滤波算法，与现有方法不同，该方法预测的集合为连续序列，便于高效恢复与调试。我们在多种智能体和数据集上验证了理论保证，证明可精确定位错误，进而利用预测集实现系统自我修复。整体方法对模型无依赖，为多智能体系统提供可解释的不确定性层。代码已开源：https://github.com/layer6ai-labs/conformal-agent-error-attribution。
原文摘要 · Abstract (English)
When multi-agent systems (MAS) fail, identifying where the decisive error occurred is the first step for automated recovery to an earlier state. Error attribution remains a fundamental challenge due to the long interaction traces that large language model-based MAS generate. This paper presents a framework for error attribution based on conformal prediction (CP) which provides finite-sample, distribution-free coverage guarantees. We introduce new algorithms for filtration-based CP designed for sequential data such as agent trajectories. Unlike existing CP algorithms, our approach predicts sets that are contiguous sequences to enable efficient recovery and debugging. We verify our theoretical guarantees on a variety of agents and datasets, show that errors can be precisely isolated, then use prediction sets to rollback MAS to correct their own errors. Our overall approach is model-agnostic, and offers a principled uncertainty layer for MAS error attribution. We release code at https://github.com/layer6ai-labs/conformal-agent-error-attribution.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。