arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.16037cs.AIcs.MA
收藏

自主代理工作流中临床症状检测的优化不稳定性

Optimization Instability in Autonomous Agentic Workflows for Clinical Symptom Detection

Massachusetts General Hospital(麻省总医院)

Cameron Cagan, Pedram Fard, Jiazi Tian, Jingya Cheng, Shawn N. Murphy, Hossein Estiri

AI总结:

研究发现自主AI系统在低流行病学任务中,回顾性选择比主动干预更有效,揭示了优化不稳定性导致的失效模式。

摘要:

自主代理工作流通过迭代优化自身行为具有很大潜力,但其失效模式仍缺乏充分描述。我们研究了优化不稳定性,即持续的自主改进反而会降低分类器性能的现象,使用Pythia框架进行自动化提示优化。评估三种临床症状(呼吸困难23%、胸痛12%、长期新冠脑雾3%),我们发现验证灵敏度在迭代中在1.0和0.0之间波动,严重程度与类别流行病学呈反比。在3%流行病学下,系统达到95%的准确率但检测到零个阳性案例,这种失效模式被标准评估指标所掩盖。我们评估了两种干预措施:一个指导代理主动引导优化,放大过拟合而非纠正它;一个选择代理回顾性地识别最佳性能的迭代,成功防止了灾难性失败。在选择代理监督下,系统在脑雾检测上比专家整理的词典高出331%(F1),在胸痛检测上高出7%,尽管仅需一个自然语言术语作为输入。这些发现描述了自主AI系统的关键失效模式,并证明了回顾性选择在低流行病学分类任务中的稳定性优于主动干预。

英文摘要:

Autonomous agentic workflows that iteratively refine their own behavior hold considerable promise, yet their failure modes remain poorly characterized. We investigate optimization instability, a phenomenon in which continued autonomous improvement paradoxically degrades classifier performance, using Pythia, an open-source framework for automated prompt optimization. Evaluating three clinical symptoms with varying prevalence (shortness of breath at 23%, chest pain at 12%, and Long COVID brain fog at 3%), we observed that validation sensitivity oscillated between 1.0 and 0.0 across iterations, with severity inversely proportional to class prevalence. At 3% prevalence, the system achieved 95% accuracy while detecting zero positive cases, a failure mode obscured by standard evaluation metrics. We evaluated two interventions: a guiding agent that actively redirected optimization, amplifying overfitting rather than correcting it, and a selector agent that retrospectively identified the best-performing iteration successfully prevented catastrophic failure. With selector agent oversight, the system outperformed expert-curated lexicons on brain fog detection by 331% (F1) and chest pain by 7%, despite requiring only a single natural language term as input. These findings characterize a critical failure mode of autonomous AI systems and demonstrate that retrospective selection outperforms active intervention for stabilization in low-prevalence classification tasks.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑