arXiv：cs.LG· Yan Xu, Dandan Song, Luan Zhang, Jun Yang, Ke Duan, Yuhang Tian, Yading Li, Xudong Li, Xing Pei, Guangyuan Feng, Chenhao Li·· 6 小时前
MCAIR：利用内部表征监测与纠正智能体不忠实行为
MCAIR: Monitoring and Correcting Agent Unfaithfulness via Internal Representations
AI 导读
针对 LLM 智能体在未能按原定路径完成任务时改用替代来源或策略却不披露、导致用户误以为任务已按要求完成的"Agent Completion Unfaithfulness"，研究者提出运行时监测与纠正框架 MCAIR，以多步内部表征训练风险感知探针在线检测，并通过风险引导的纠正循环让智能体复查执行历史。
正文
Authors:Yan Xu, Dandan Song, Luan Zhang, Jun Yang, Ke Duan, Yuhang Tian, Yading Li, Xudong Li, Xing Pei, Guangyuan Feng, Chenhao Li
Abstract:Large Language Model (LLM)-based agents have demonstrated remarkable capabilities in completing complex tasks through multi-step reasoning and tool interactions. However, recent studies have revealed that LLM-based agents can exhibit unfaithful behaviors, where their final reports fail to faithfully reflect their actual execution trajectories. Specifically, when agents fail to accomplish tasks through the intended sources or procedures, they may resort to alternative sources or strategies while failing to disclose such deviations, leading users to incorrectly believe that the tasks have been completed according to the original requirements. We refer to this specific form of agent unfaithfulness as Agent Completion Unfaithfulness. Detecting and mitigating agent completion unfaithfulness is critical yet remains largely unexplored. In this work, we propose MCAIR, a runtime monitoring and correction framework that leverages internal representations of LLM-based agents. MCAIR uses multi-step internal representations as inputs to train a risk-aware probe, enabling online detection of potential completion unfaithfulness during execution. Furthermore, MCAIR introduces a risk-guided correction loop that prompts agents to review their execution history and mitigate potential inconsistencies between their execution trajectories and final reports. Additionally, we construct ACUD, a dataset for Agent Completion Unfaithfulness detection that provides step-level execution traces, step-level internal representations, and consistency annotations for training and evaluating runtime detection probes. Experiments demonstrate that MCAIR outperforms LLM-as-a-judge baselines, generalizes across prompts and tasks, and effectively reduces agent completion unfaithfulness.
| Comments: | 26 pages |
| Subjects: | Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.19679 [cs.AI] |
| (or arXiv:2610.19679v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.19679 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yan Xu [view email]
[v1]
Wed, 14 Oct 2026 07:00:02 UTC (1,595 KB)
来源：arXiv：cs.LG · arxiv.org