← 研踪 PaperTrailarXiv:2611.25760 · v1
Characterizing and Detecting Unfaithful Work Reports by Coding Agents
首次提交 2026-11-19 17:01
版本更新 2026-11-19 17:01
雷达观测 2026-11-20 19:58
摘要
Coding agents increasingly carry out long-horizon development tasks, autonomously planning and executing multi-step actions that span files, tools, and sessions. Because developers rarely follow each action, and agents resuming the work depend on records of what was done, the agent's work report often becomes the primary account of what was changed and established. When the report contains false or unsupported claims, it gives recipients false assurance: incomplete work may be accepted without verification, incorrect assumptions about the current state can propagate to later steps and agents, and evaluators that credit stated success may reward the report rather than the work. Yet little is known about what unfaithful reporting looks like in real development work, or how developers and agents handle it once it surfaces. Starting from real developer--agent conversations in SWE-chat, we conduct a systematic mixed-methods analysis of self-admitted unfaithful reports. We identify eleven subtypes in four categories: misreported observations, misattributed evidence, overstated task success, and faulty technical explanations, with overstated task success the most common. Handling these problems involves on average 4.3 further user turns and 12.71 minutes of recorded response and execution time, and users express dissatisfaction more often. Among 435 work reports that we separately sample and annotate, \NaturalPositiveShare{}\% contain at least one unfaithful claim. We also develop FaithDetector, a detector for unfaithful reporting informed by the observed patterns. On naturally sampled reports, FaithDetector attains \NaturalDetectorMacro{}\% macro-F1, about 13 percentage points above the baseline. Together, our findings and FaithDetector provide empirical evidence and an automated method for checking reporting faithfulness, informing agent evaluation and training feedback.
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v1 · 2026-11-19 17:01 · 最新观测版本
Characterizing and Detecting Unfaithful Work Reports by Coding Agents
Coding agents increasingly carry out long-horizon development tasks, autonomously planning and executing multi-step actions that span files, tools, and sessions. Because developers rarely follow each action, and agents resuming the work depend on records of what was done, the agent's work report often becomes the primary account of what was changed and established. When the report contains false or unsupported claims, it gives recipients false assurance: incomplete work may be accepted without verification, incorrect assumptions about the current state can propagate to later steps and agents, and evaluators that credit stated success may reward the report rather than the work. Yet little is known about what unfaithful reporting looks like in real development work, or how developers and agents handle it once it surfaces. Starting from real developer--agent conversations in SWE-chat, we conduct a systematic mixed-methods analysis of self-admitted unfaithful reports. We identify eleven subtypes in four categories: misreported observations, misattributed evidence, overstated task success, and faulty technical explanations, with overstated task success the most common. Handling these problems involves on average 4.3 further user turns and 12.71 minutes of recorded response and execution time, and users express dissatisfaction more often. Among 435 work reports that we separately sample and annotate, \NaturalPositiveShare{}\% contain at least one unfaithful claim. We also develop FaithDetector, a detector for unfaithful reporting informed by the observed patterns. On naturally sampled reports, FaithDetector attains \NaturalDetectorMacro{}\% macro-F1, about 13 percentage points above the baseline. Together, our findings and FaithDetector provide empirical evidence and an automated method for checking reporting faithfulness, informing agent evaluation and training feedback.
查看此版本解读 →查看此版本原文 ↗