Metraipolis › archive › dc_cd6fd6ba8bb6

When Tools Lie: Reliability of Mathematical Agents Under Corrupted Tool Feedback — arXiv:2610.08097

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.08097
License (as archived)
CC0-1.0
Archived
2026-10-07T11:21:28.673Z
SHA-256 of the archived text
1e0cd7344200613e772905db6da91a3276067376b106e2ae4600258db263250f
Stable URL of this page
https://metraipolis.com/documento/dc_cd6fd6ba8bb6

How to cite

When Tools Lie: Reliability of Mathematical Agents Under Corrupted Tool Feedback — arXiv:2610.08097 — arXiv (en), https://arxiv.org/abs/2610.08097 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cd6fd6ba8bb6 (Metraipolis archive dc_cd6fd6ba8bb6, sha256 1e0cd7344200613e772905db6da91a3276067376b106e2ae4600258db263250f)
BibTeX
@misc{dccd6fd6ba8bb6,
  title = {When Tools Lie: Reliability of Mathematical Agents Under Corrupted Tool Feedback — arXiv:2610.08097},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.08097},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cd6fd6ba8bb6; sha256 1e0cd7344200613e772905db6da91a3276067376b106e2ae4600258db263250f},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.08097. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Mathematical problem solving often requires deterministic computational steps that agents delegate to tools and implicitly trust. Yet tools can fail silently, returning plausible but incorrect results. How well can agents detect and correct corrupted tool call outputs? We study this through a controlled corruption framework where a hidden interceptor replaces tool call results with plausible incorrect information on targeted problems. We evaluate agents across 31 problems under four verification designs including no verification (baseline), mandatory same-context reflection, optional fresh-context verification, and optional structural verification. Without verification, corruption causes dramatic accuracy loss, from 100% down to 72.4%. Mandatory reflection fully recovers this performance to 100%. Optional verification improves accuracy only when models actively invoke it. Our results show that checking frequency is strongly associated with robustness differences, while unequal invocation prevents a controlled comparison of verifier quality. A supporting recovery experiment shows that full problem restart succeeds in 100% of cases after explicit detection. These findings demonstrate that verifier availability and verification policy are separate components of mathematical-agent reliability. Mandatory policies enforce verification while optional policies depend on the model's own choice to invoke it.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI); Software Engineering (cs.SE).
Pagina del lavoro: https://arxiv.org/abs/2610.08097 — 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 2026-10-07T11:15:24.796Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).