Metraipolis › archive › dc_fa829621c418

Localizing Emergent Failures in Agentic AI: Recovering Minimal Repair Families via Counterfactual Replay — arXiv:2608.29228

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.29228
License (as archived)
CC0-1.0
Archived
2026-10-17T07:41:00.773Z
SHA-256 of the archived text
e07b0c76389d5b9cdc443052b18260156a9dba4e90a35e52bc16a2b05e5201af
Stable URL of this page
https://metraipolis.com/documento/dc_fa829621c418

How to cite

Localizing Emergent Failures in Agentic AI: Recovering Minimal Repair Families via Counterfactual Replay — arXiv:2608.29228 — arXiv (en), https://arxiv.org/abs/2608.29228 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fa829621c418 (Metraipolis archive dc_fa829621c418, sha256 e07b0c76389d5b9cdc443052b18260156a9dba4e90a35e52bc16a2b05e5201af)
BibTeX
@misc{dcfa829621c418,
  title = {Localizing Emergent Failures in Agentic AI: Recovering Minimal Repair Families via Counterfactual Replay — arXiv:2608.29228},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.29228},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fa829621c418; sha256 e07b0c76389d5b9cdc443052b18260156a9dba4e90a35e52bc16a2b05e5201af},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.29228. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Failures in agentic AI systems can arise from interactions among messages exchanged by multiple large language model (LLM) agents. Pointwise attribution cannot distinguish a jointly necessary repair from alternative singleton repairs. We formulate Minimal Repair Family Recovery (MRFR): recovering all inclusion-minimal event sets whose counterfactual replay restores task success within a declared size bound. We propose Graph-Constrained Joint Replay (GCJR), which slices failure-relevant events from an execution dependency graph, constructs graph-feasible singleton and pair candidates, and verifies them by replay with paired clean counterparts. For fixed replay outcomes, GCJR is exact within its declared graph domain. On 90 in-scope cases from a 120-DAG controlled benchmark, GCJR achieves 1.000 Family Exact Match while reducing mean replay calls from 56.3 to 25.3 (55.1%) relative to exhaustive search. On a 24-case, four-agent LLM pilot, it again achieves 1.000 Family Exact Match and reduces mean model calls from 21.0 to 10.0 (52.4%); single-event replay misses jointly necessary repairs.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-01 in cs.MA).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Multiagent Systems (cs.MA).
Pagina del lavoro: https://arxiv.org/abs/2608.29228 — 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 2026-10-17T07:27:41.995Z dalla pagina https://arxiv.org/catchup/cs.MA/2026-09-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).