Metraipolis › archive › dc_1e657e62c2da
LongRCA Bench: Root-Cause Localization in Long-Horizon Agent Trajectories — arXiv:2608.15242
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.15242
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T17:13:45.138Z
- SHA-256 of the archived text
1e12e2c657dd1192561a2fdc878fe8e08768eaeb42af7c5c8131c398dc92c6ff- Stable URL of this page
https://metraipolis.com/documento/dc_1e657e62c2da
How to cite
LongRCA Bench: Root-Cause Localization in Long-Horizon Agent Trajectories — arXiv:2608.15242 — arXiv (en), https://arxiv.org/abs/2608.15242 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_1e657e62c2da (Metraipolis archive dc_1e657e62c2da, sha256 1e12e2c657dd1192561a2fdc878fe8e08768eaeb42af7c5c8131c398dc92c6ff)
BibTeX
@misc{dc1e657e62c2da,
title = {LongRCA Bench: Root-Cause Localization in Long-Horizon Agent Trajectories — arXiv:2608.15242},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.15242},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_1e657e62c2da; sha256 1e12e2c657dd1192561a2fdc878fe8e08768eaeb42af7c5c8131c398dc92c6ff},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.15242. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
In long agent executions, an early error can persist through later actions and checks, while evidence needed to trace its origin is dispersed across the history. Short histories offer limited tests of recovering error origins across substantial subsequent execution. We introduce LongRCA Bench: 1,140 complete failed trajectories from five sources, all human-annotated for responsible roles and earliest decisive root-cause steps. Reference roots precede completion by a median of 48 recorded steps; 28.4% of trajectories contain over 100 subsequent steps. With DeepSeek-V4-Flash on the full benchmark, the strongest of five evaluated baselines achieves 13.2% exact root-step accuracy. We propose Root-Cause Trajectory Attribution (RCTA), a training-free method that organizes original candidate records and explicit handoff instructions for attribution. Segment summaries and a trajectory outline guide candidate retrieval; available handoff records supply upstream instruction context for the final instruction-execution comparison. With the same backbone and scoring protocol, RCTA reaches 24.1% exact root-step accuracy and 51.1% responsible-role accuracy. LongRCA-Mini provides 200 fixed trajectories for lower-cost comparative screening. Even with RCTA, fewer than one quarter of reference roots are recovered exactly. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-18 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI) ; Software Engineering (cs.SE). Pagina del lavoro: https://arxiv.org/abs/2608.15242 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-18T16:29:13.779Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).