Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Conformal Agent Error Attribution

About

When multi-agent systems (MAS) fail, identifying where the decisive error occurred is the first step for automated recovery to an earlier state. Error attribution remains a fundamental challenge due to the long interaction traces that large language model-based MAS generate. This paper presents a framework for error attribution based on conformal prediction (CP) which provides finite-sample, distribution-free coverage guarantees. We introduce new algorithms for filtration-based CP designed for sequential data such as agent trajectories. Unlike existing CP algorithms, our approach predicts sets that are contiguous sequences to enable efficient recovery and debugging. We verify our theoretical guarantees on a variety of agents and datasets, show that errors can be precisely isolated, then use prediction sets to rollback MAS to correct their own errors. Our overall approach is model-agnostic, and offers a principled uncertainty layer for MAS error attribution. We release code at https://github.com/layer6ai-labs/conformal-agent-error-attribution.

Naihe Feng, Yi Sui, Shiyi Hou, Ga Wu, Jesse C. Cresswell• 2026

Related benchmarks

TaskDatasetResultRank
Error AttributionGSM8k Right Dense
Removal Rate64
30
Error AttributionGSM8k Mid Dense
Removal Rate63
30
Error AttributionGSM8k Left Dense
Removal Rate71
30
Error AttributionMATH
Removal Rate29
30
Error AttributionWho&When--
30
Step-level error discriminationMATH and GSM8k (test)--
4
Automated agent rollbackGSM8K Left Dense variant (test)
Success Rate77
3
Automated agent rollbackGSM8K Mid Dense variant (test)
Success Rate68
3
Automated agent rollbackGSM8K Right Dense variant (test)
Success Rate75
3
Showing 9 of 9 rows

Other info

GitHub

Follow for update