修复了情绪监控误报问题，发现时钟校准的监控在真实任务中无法准确捕捉关键时刻。
Bistable by Construction: Wall-Clock-Calibrated State Monitors Have No Moment-Detection Regime at Agent Cadence
- 用墙钟时间校准的累积器会因动作间隔变化产生误报或失灵
- 在1-30秒动作间隔内报警率高达20/20次，超过60秒则完全沉默
- 适合关注实时监控机制可靠性的系统设计者和安全评估人员
自主代理的运行时监控常通过阈值触发累积状态（如行为基线、漂移统计或建模的情绪状态）。此前报告的‘状态饱和陷阱’显示，在SWE-bench调试代理上，基于情绪引擎的阈值触发导致持续报警（Modgil 2026）。事后审计发现，该引擎在动作间接受到的dt=0，其指数衰减未生效，因此原陷阱实为纯累加结果。本文更正记录（勘误，v2），将此缺陷视为实验。关键变量是监控动态的校准方式：采样时间（如CUSUM）还是墙钟时间（如半衰期以秒计）。固定速率流下二者一致，但在代理流中，因动作间隔可差数个数量级，二者不等。对20条轨迹在均匀间隔dt∈{0..600}s的预注册扫描表明，墙钟级触发器存在两个阶段：当dt≤1s时持续报警（20/20；中位18次）；当dt≥60s时完全沉默。所有临界值均落在(1,30]秒区间。真实代理运行的延迟中位数为1.53秒（第90百分位2.33秒），处于陷阱区间，验证了修正机制下的经验发现。该结构属于校准类别属性：仅用原始误差流的最小墙钟累加器即可复现同一悬崖效应，而同一数据流上的采样时间CUSUM则严格保持dt不变（20/20）。带滞后性的上升沿触发器在每种条件下仅触发0-3次。结论：墙钟校准的漏桶型监控在代理流中不存在作为时刻检测器的任何有效区间；过渡检测虽能逃出陷阱，但无法恢复人类干预时机。
原文摘要 · Abstract (English)
Runtime monitors for autonomous agents commonly threshold an accumulated internal state - a behavioural baseline, a drift statistic, or, in our prior work, a modelled affective state. We previously reported a State Saturation Trap: threshold-on-state triggers over a continuous affect engine become near-constant alarms on SWE-bench debugging agents (Modgil 2026). A post-release audit found the engine received dt=0 between actions, so its exponential decay never operated: the published trap is a pure-accumulator result. We correct the record (erratum, v2) and treat the flaw as an experiment. The key variable it exposes is whether a monitor's dynamics are calibrated in sample time (per observation, as in CUSUM) or wall-clock time (half-lives in seconds, as in affect models and EMA baselines). On fixed-rate streams these coincide; on agent streams, where inter-action time varies by orders of magnitude, they do not. A pre-registered sweep over uniform intervals (dt in {0..600}s) on 20 trajectories shows the wall-clock level trigger has two regimes: at dt<=1s a constant alarm (20/20; median 18 firings); at dt>=60s silent. Every critical dt lies in (1,30]s. Real agent runs measure latency at median 1.53s (p90 2.33s); real coding cadence sits inside the trap regime, vindicating the empirical finding under a corrected mechanism. The structure is a property of the calibration class, not the engine: a minimal wall-clock accumulator over the raw error stream reproduces the same cliff, while a sample-time CUSUM over the identical stream is exactly dt-invariant (20/20). A rising-edge trigger with hysteresis fires 0-3 times per trajectory in every condition. We conclude that wall-clock-calibrated leaky-integrator monitors admit no regime in which they act as moment detectors on agent streams; transition detection escapes the trap at every cadence, but does not recover human intervention timing.
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。