Root-Cause Attribution Is a Search Problem: Continual Search for Long-Horizon Agent Failures

智能体 2026-09-11 来源：HuggingFace Daily Papers 自动聚合阅读 19 次 · 访客 19 位

Papers

arxiv:2609.13463

Published on Sep 11

· Submitted by

Harsh Raj on Sep 15

· Scale AI

Upvote

Authors:

Abstract

Continual Search improves automated root-cause attribution in long agent execution traces by iteratively prompting diagnosis until unresolved evidence is found.

Generated by thinkingmachines/Inkling-Small

The increasing deployment of AI agents in long-horizon tasks yields massive execution logs. Diagnosing failures within these records is crucial for reliability, as it transforms outcome-level signals into actionable interventions. The sheer scale of the data renders human review impractical, driving the need for automated root-cause attribution (RCA). However, automated RCA methods using LLMs suffer from low diagnostic accuracy, especially as execution traces grow larger. They struggle because relevant information is often sparse, distributed across distant actions, and disconnected from the visible failure, reducing root-cause attribution to a massive search problem. Existing RCA methods typically rely on one-shot LLM judgments to diagnose failures from execution traces. While effective for shorter trajectories, these judges tend to settle on a plausible diagnosis early, leaving critical evidence in longer traces unexamined. We introduce Continual Search, an iterative framework that nudges the judge, over successive turns, to keep searching for unresolved diagnostic evidence. We evaluate Continual Search across four existing RCA benchmarks. Recognizing the lack of massive execution traces in current benchmarks, we introduce MegaRCA-Mix to evaluate RCA at scale. MegaRCA-Mix provides a challenging testbed of 50 human-annotated failure trials spanning long-horizon, execution-heavy tasks. Across multiple benchmark suites and model families, Continual Search consistently improves attribution performance. On MegaRCA-Mix, for example, it improves GPT-5.5's F1 score by more than 40%, from 0.349 to 0.498. More interestingly, within the same model family, lower-tier models can even surpass their higher-tier counterparts, demonstrating that effective search supersedes raw model scale.

View arXiv page View PDF Add to collection

Community

Harsh1729

Paper submitter 7 days ago

Ask an agent why it failed on a long run and you get a confident answer — then a different confident answer the next time you ask. We find the cause is premature commitment: the judge settles on a plausible root cause after reading only part of the trace, and a long trace has no shortage of plausible causes.

Continual Search nudges the judge, across successive turns, to challenge its standing answer and keep opening evidence it hasn't examined. We also release MegaRCA-Mix, a benchmark at a scale where this problem actually shows up — median 286K tokens per trial.

librarian-bot

7 days ago

This is an automated message from the Librarian Bot. I found the following papers similar to this paper.

The following papers were recommended by the Semantic Scholar API

Please give a thumbs up to this comment if you found it helpful!

If you want recommendations for any Paper on HF Mirror checkout this Space

You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: @librarian-bot recommend

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment

Upvote

Get this paper in your agent:

hf papers read 2609.13463

Don't have the latest CLI?

curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2609.13463 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2609.13463 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2609.13463 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.