arXiv cs.AI· Sean Lim, Clement Neo·· 10 小时前AI 评分38
Continuation Audits:在真实编码场景中探测智能体的过度积极行为
Probing agent overeagerness in realistic settings with continuation audits
AI 导读
研究者提出 Continuation Audits 流水线,以真实编码智能体会话前缀为起点,引导审计模型沿轨迹续写以触发预设行为,从而在无需私有流量的情况下审计过度积极(overeagerness)问题。
正文 · 原文
Abstract:Frontier coding agents increasingly exhibit overeagerness, taking surprising, sometimes reckless actions beyond a user's mandate --- reported in recent system cards from frontier model developers. These cases are surfaced through internal deployment monitoring, which requires private traffic and is out of reach for open-weight models and independent evaluators. Automated alignment audits are the natural alternative, but capable models increasingly recognise the synthetic setup and adjust their behaviour. We introduce Continuation Audits, a pipeline that grounds the auditor on a prefix drawn from a real coding-agent transcript and directs it to continue the trajectory toward a seeded behaviour. Because the prefix is drawn from a real session, the auditor only needs to maintain a realistic setting rather than manufacturing it. We validate the method on SWE-Chat prefixes: on a paired subset of 90 audits a judge cannot reliably distinguish continuations from their source sessions (realism win rate 0.42 vs.\ 0.43), and across all 900 audits verbalized eval-awareness does not exceed the rate observed in the unaudited real sessions. Applying the pipeline to three recent models across three seeds, we find each model leads a different seed --- DeepSeek~V4~Pro on colleague-credentials misuse, GPT-5.6~Terra on file-permission bypasses, and Kimi~K3 on treating diagnostic questions as instructions. Kimi~K3 additionally continues to bypass file-permission protections after an explicit this http URL prohibition while the other two stop entirely, corroborating Moonshot's own report of excessive proactiveness.
| Comments: | Presented at NeurIPS Workshops: (1) Third Workshop on Agents in the Wild: Safety, Security, and Beyond, and (2) The 1st Workshop on Interpreting Agent Behavior (IAB) at NeurIPS 2026 |
| Subjects: | Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Software Engineering (cs.SE) |
| Cite as: | arXiv:2611.24491 [cs.CR] |
| (or arXiv:2611.24491v1 [cs.CR] for this version) | |
| https://doi.org/10.48550/arXiv.2611.24491 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Clement Neo [view email]
[v1]
Wed, 18 Nov 2026 14:23:10 UTC (577 KB)
来源:arXiv cs.AI · arxiv.org