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.
Internet of Agentic AI: Incentive-Compatible Distributed Teaming and Workflow
Department of Electrical and Computer Engineering, Tandon School of Engineering, New York University, Brooklyn, NY, USA(电气与计算机工程系,工程学院,纽约大学,布鲁克林,纽约,美国)
Composing Verifiable Conceptual Models via Building Blocks: Towards Design-Time Verification of Agentic AI Workflows
Team EVERGREEN Inria Centre Inria d’Université Côte d’Azur(法国国家信息与自动化研究所蔚蓝海岸大学中心EVERGREEN团队);Department of Data Science William & Mary(威廉与玛丽学院数据科学系);Office of Enterprise Research and Innovation Old Dominion University(欧道明大学企业研究与创新办公室)
University of Sydney(悉尼大学);University of Adelaide(阿德莱德大学);Mohamed bin Zayed University of Artificial Intelligence(穆罕默德·本·扎耶德人工智能大学);Carnegie Mellon University(卡内基梅隆大学)
A Practical Guide for Designing, Developing, and Deploying Production-Grade Agentic AI Workflows
Old Dominion University(旧 Dominion 大学);Florida International University(佛罗里达国际大学);Nanyang Technological University(南洋理工大学);University of Colombo(科伦坡大学)
A Sober Look at Agentic Misalignment in Automated Workflows
University of Virginia(弗吉尼亚大学);Georgia Institute of Technology(佐治亚理工学院);University of Utah(犹他大学);University of Notre Dame(圣约翰大学);George Washington University(乔治华盛顿大学)
Agentic workflow enables the recovery of critical materials from complex feedstocks via selective precipitation
Physical and Computational Sciences Directorate, Pacific Northwest National Laboratory(太平洋西北国家实验室物理与计算科学部门);Department of Chemical Engineering, University of Washington(华盛顿大学化学工程系)
Learning to Compose for Cross-domain Agentic Workflow Generation
Hong Kong University of Science and Technology(香港理工大学);Hong Kong University of Science and Technology (Guangzhou)(香港理工大学(广州));Southeast University(东南大学);University of Tsukuba(茨口大学)
HyEvo: Self-Evolving Hybrid Agentic Workflows for Efficient Reasoning
East China Normal University(华东师范大学);Beihang University(北京航空航天大学);Shanghai University of International Business and Economics(上海对外经贸大学);Fudan University(复旦大学)
Canadian University Dubai(迪拜加拿大大学);Mohamed bin Zayed University of Artificial Intelligence(穆罕默德·本·扎耶德人工智能大学);Zayed University(扎耶德大学);Hong Kong Baptist University(香港浸会大学);University of Technology Sydney(悉尼科技大学)
Plausible but Wrong: A case study on Agentic Failures in Astrophysical Workflows
Bonn-Aachen International Center for Information Technology, University of Bonn, Germany(波恩-阿森国际信息科技中心,波恩大学,德国);Lamarr Institute for Machine Learning and Artificial Intelligence, Germany(拉马尔机器学习与人工智能研究所,德国)