# SearchAuditor: Auditing and Attributing Failures in Long-Horizon Search Agents

arXiv: https://arxiv.org/abs/2608.05212 · HTML: https://mlsanity.com/paper/2608.05212/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Web, search, and research agents, one of 16 directions in LLM agents and tool use.
- Builds on 14 of its 55 references that are in the map.
- First posted on arXiv 5 Aug 2026.

## Placed in

- LLM agents and tool use / [Web, search, and research agents](https://mlsanity.com/agents/web-search-and-research-agents/)

## Builds on (most cited first, 14 of 14)

- [WebGPT: Browser-assisted question-answering with human feedback](https://mlsanity.com/paper/2112.09332/) (2021)
- [BrowseComp: A Simple Yet Challenging Benchmark for Browsing Agents](https://mlsanity.com/paper/2504.12516/) (2025)
- [BrowseComp-ZH: Benchmarking Web Browsing Ability of Large Language Models in Chinese](https://mlsanity.com/paper/2504.19314/) (2025)
- [TRAIL: Trace Reasoning and Agentic Issue Localization](https://mlsanity.com/paper/2505.08638/) (2025)
- [SealQA: Raising the Bar for Reasoning in Search-Augmented Language Models](https://mlsanity.com/paper/2506.01062/) (2025)
- [OpenSeeker: Democratizing Frontier Search Agents by Fully Open-Sourcing Training Data](https://mlsanity.com/paper/2603.15594/) (2026)
- [OpenResearcher: A Fully Open Pipeline for Long-Horizon Deep Research Trajectory Synthesis](https://mlsanity.com/paper/2603.20278/) (2026)
- [Failure as a Process: An Anatomy of CLI Coding Agent Trajectories](https://mlsanity.com/paper/2607.09510/) (2026)
- [QUEST: Training Frontier Deep Research Agents with Fully Synthetic Tasks](https://mlsanity.com/paper/2605.24218/) (2026)
- [FALAT: Tracing Failures in LLM Agent Trajectories via Dependency-Guided Search](https://mlsanity.com/paper/2606.00765/) (2026)
- [TrajAudit: Automated Failure Diagnosis for Agentic Coding Systems](https://mlsanity.com/paper/2605.26563/) (2026)
- [MemTrace: Tracing and Attributing Errors in Large Language Model Memory Systems](https://mlsanity.com/paper/2605.28732/) (2026)
- [Why Your Deep Research Agent Fails? On Hallucination Evaluation in Full Research Trajectory](https://mlsanity.com/paper/2601.22984/) (2026)
- [LiveBrowseComp: Are Search Agents Searching, or Just Verifying What They Already Know?](https://mlsanity.com/paper/2605.28721/) (2026)

## Built on by (most cited first, 2 of 2)

- [Where Scientific Search Agents Fail: Decision-Checkpoint Auditing of Exposure and Inspection Attempts](https://mlsanity.com/paper/2609.38670/) (2026)
- [Avoiding What Your Teacher Already Knows: Entity Familiarity for Deep-Search Distillation](https://mlsanity.com/paper/2610.16251/) (2026)

## Abstract (arXiv)

Deep search agents tackle challenging questions through long-horizon web interactions, a process that is both complex and fragile: small reasoning errors may propagate through long, noisy trajectories into fluent but incorrect answers. Diagnosing such failures is difficult, requiring the manual inspection of extremely long execution traces, which could be beyond human capacity. We therefore introduce SearchAuditBench, a benchmark that evaluates whether LLM auditors can localize, attribute, and repair these failures, thereby reducing the human burden. SearchAuditBench comprises 1,243 failed trajectories, averaging 73.1 messages and 65.1K tokens, collected from eight open-weight models on five deep-search benchmarks, each expert-annotated with the critical error step, a search-specific root cause, and a reference repair with grading rubrics. We further propose SearchAuditor, a multi-perspective auditing framework that effectively localizes, attributes, and repairs search-agent failures through evidence-grounded adjudication. Experimental results show that even the strongest baseline, when powered by a frontier model like GPT-5.5, attains only a 26.6% end-to-end pass rate. In contrast, our SearchAuditor consistently outperforms all baselines across different frontier models, achieving an end-to-end pass rate of 32.3%, and resuming failed runs with its repairs enables agents to better recover from errors.