task · first seen 5 Oct 2026
failure attribution in LLM agents
Also: failure attribution, root-cause attribution, agent failure attribution, agent failure localization, automated failure attribution, fault localization in multi-agent systems, failure attribution in multi-agent systems
Identifying the step in an agent trajectory that causes task failure, using judges or counterfactual restarts.
Mentions per week
PapersCodeDiscussion
Earliest sources
- Where Did the Repair First Go Wrong? Localizing the Origins of Silent Failures in Agentic Vulnerability RepairPapers · 5 Oct 2026 · Wenji Bai, Muhammad Waseem, Zeeshan Rasheed, Jaakko Peltonen, Pekka Abrahamsson
- ReCast: Attribution-Oriented Step Representation Learning for LLM-Based Agent SystemsPapers · 8 Oct 2026 · Weilin Jin, Mingyu Wang, Taiyu Zhu, Ziqi Zhou, Wenbo Li, Haoyang Huang, Nan Duan…
- Error-Propagation Modeling for Failure Attribution in LLM-Based Multi-Agent SystemsPapers · 8 Oct 2026 · Jiaqi Liao, Yuanzhao Zhai, Huanxi Liu, Xu Zhang, Zheming Zhuang, Dawei Feng, Bo…
- Can Failed Trajectories Reveal Where Agents Go Wrong? Agentic Failure Attribution Without Step-Level LabelsPapers · 9 Oct 2026 · Do Nguyen-Thanh, Junlin Fang, Bing Li, Zhen Fang, Sean Du
- Structural Baselines Matter for Agent Failure Detection and AttributionPapers · 12 Oct 2026 · Ramesh Arvind Naagarajan, Stefan Streif
- Executable Prefix Values for Locating Agent FailuresPapers · 13 Oct 2026 · Bingyu Shen, Jiaqi Tang, Lan Wei, Yi Sheng, J. Jenny Li, Boyang Li
Latest sources
- CoLFA: Collaboration-Guided Lightweight Failure Attribution for LLM-Based Multi-Agent SystemsPapers · 11 Nov 2026 · Yufei Luo, Yuchen Ling, Chunrong Fang
- Dependable Long-Horizon LLM Agents: A Trajectory-Centered Empirical Study of Where Errors Enter and Which Controls Reach ThemPapers · 9 Nov 2026 · Shaohuang Wang, Jinghao Wen, Xulei Zhang, Xun Jiao
- Attributing Agent Failures to Their History Needs an Executed Counterfactual: Toward a Reporting StandardPapers · 3 Nov 2026 · Egor Pakhomov, Erik Nijkamp
- Trial by Trace: Benchmarking AI Agent Harnesses for Reliability and Failure AttributionPapers · 23 Oct 2026 · Zhiyong Cao, Hengjie Zhu, Yongxin Wang, Xuchen Li, Jintao Li, Shuxia Liu, Dunqia…
- A Mechanistic Study of Automated Failure Attribution in LLM-based Agentic SystemsPapers · 20 Oct 2026 · Wendy Zheng, Liang Wu, Yinhan He, Mayank Darbari, Jundong Li, Liangjie Hong
- What Would Removing This Corruption Change? Failure Attribution for Vision-Language-Action ModelsPapers · 19 Oct 2026 · Carson Sobolewski, Anjali Parashar, Fei Chen, Yingke Li, Chuchu Fan
- From Trace to Fix: Residual Task State in Long-Horizon AgentsPapers · 19 Oct 2026 · Zile Huang, Ser-nam Lim
- IF-MAP: Influence-Function-Inspired Agent Prioritization in Multi-Agent SystemsPapers · 17 Oct 2026 · Jie Mao, Pingchen Lu, Zikun Qu, Jiahang Sun, Xiangyi Wang, Zhiwei Shang, Zhiyong…
- BLADE: Auditing Information Use for Failure Attribution in LLM Multi-Agent SystemsPapers · 13 Oct 2026 · Xiaoguang Guo, Yaohao Long, Chuxu Zhang
- Auditing Failure Attribution: Label Predictability and Judge BehaviorPapers · 13 Oct 2026 · Tianyang Xu, Jing Gao
- Executable Prefix Values for Locating Agent FailuresPapers · 13 Oct 2026 · Bingyu Shen, Jiaqi Tang, Lan Wei, Yi Sheng, J. Jenny Li, Boyang Li
- Structural Baselines Matter for Agent Failure Detection and AttributionPapers · 12 Oct 2026 · Ramesh Arvind Naagarajan, Stefan Streif