paper-with-me

홈 › Papers

REFLECT: Intervention-Supported Error Attribution for Silent Failures in LLM Agent Traces

2026-06-08 · Xiaofeng Lin, Yingxu Wang, Tung Sum Thomas Kwok, Daniel Guo, Sahil Arun Nale, Charles Fleming, Guang Cheng arxiv

Large language model (LLM) agents now solve complex tasks through long plan-and-execution traces, yet the ability to locate errors in a completed traces still lags far behind, especially in the \emph{silent failure} regime. Existing approaches predict suspect steps via classifiers or LLM judges, or recover correct answers via retry, but none feed the intervention outcome back to \emph{refine the attribution itself}. We propose \methodname, a method that closes this gap by diagnosing a candidate error step, testing it through controlled replay with a diagnosis-specific patch, and using the verified outcome flip as contrastive evidence to refine the final attribution. Across four localization benchmarks spanning multi-hop reasoning across domains, \methodname achieves the highest localization accuracy among same-auditor methods across all four benchmarks, with the largest gains on structured tool-use traces, while providing actionable localization even when ground-truth answers are unavailable.

📄 PDF Abstract BibTeX arXiv:2606.09071

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Who Earns the Safety? Intervention-Aware Quantum Predictive Control with Safety Attribution

2026-06-08 · Yifan Wang arxiv

Hard safety filters are increasingly placed downstream of learned controllers to guarantee constraint satisfaction at run time. Yet a filtered controller that never violates a constraint may still have learned nothing ab…

Loud or Silent? A Reusable Framework for Per-Modality Failure Analysis in Multimodal Clinical AI

2026-08-02 · Quang Bui, Shlok Jaiswal, Samuel Paik-Heintz, Kevin Zhou 외 hf

Multimodal clinical models are usually judged on accuracy with every modality present, but deployment removes modalities; an echocardiogram is often unavailable where an ECG is routine. Two questions then matter beyond t…

CREG: Compass Relational Evidence Graph for Characterizing Directional Structure in VLM Spatial-Reasoning Attribution

2026-03-20 · Kaizhen Tan, Yang Feng, Heqing Du arxiv

Standard attribution heatmaps show where a vision-language model (VLM) focuses, but they do not reveal whether the recovered evidence is organized by the queried spatial relation or merely reflects image layout. To addre…

Spatial Reasoning

Tools Fail: Detecting Silent Errors in Faulty Tools

2024-06-27 · Jimin Sun, So Yeon Min, Yingshan Chang, Yonatan Bisk

Tools have become a mainstay of LLMs, allowing them to retrieve knowledge not in their weights, to perform tasks on the web, and even to control robots. However, most ontologies and surveys of tool-use have assumed the c…

DoVer: Intervention-Driven Auto Debugging for LLM Multi-Agent Systems

2025-12-07 · Ming Ma, Jue Zhang, Fangkai Yang, Yu Kang 외 arxiv

Large language model (LLM)-based multi-agent systems are challenging to debug because failures often arise from long, branching interaction traces. The prevailing practice is to leverage LLMs for log-based failure locali…

Auto Debugging