Metraipolis › archive › dc_397ab8aad2ed

When Does a Second Model Help? Cross-Model Review in LLM Verification — arXiv:2610.01471

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.01471
License (as archived): CC0-1.0
Archived: 2026-10-06T12:13:55.754Z
SHA-256 of the archived text: 97a8a5fca1ce1f7f9186571b987ddcbb07c2fa63c9c1571e21ca24fabc12ef17
Stable URL of this page: https://metraipolis.com/documento/dc_397ab8aad2ed

How to cite

When Does a Second Model Help? Cross-Model Review in LLM Verification — arXiv:2610.01471 — arXiv (en), https://arxiv.org/abs/2610.01471 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_397ab8aad2ed (Metraipolis archive dc_397ab8aad2ed, sha256 97a8a5fca1ce1f7f9186571b987ddcbb07c2fa63c9c1571e21ca24fabc12ef17)

BibTeX

@misc{dc397ab8aad2ed,
  title = {When Does a Second Model Help? Cross-Model Review in LLM Verification — arXiv:2610.01471},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.01471},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_397ab8aad2ed; sha256 97a8a5fca1ce1f7f9186571b987ddcbb07c2fa63c9c1571e21ca24fabc12ef17},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.01471. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language models now generate code, documentation, and analyses, and are increasingly used to review such output. We ask when a second review by a different model helps. Building on the author's earlier preprints, which varied context, repetition, and role structure within one model, we test model independence in a controlled experiment: 30 artifacts with 150 planted errors, 10 review conditions, and 900 review sessions with three reviewer models from two developers. In this experiment, (1) a top-tier cross-model reviewer is not significantly different in F1 from same-model review in a fresh session (CCR), which does not establish equivalence; (2) the two find partly different errors (Jaccard 41.2%); and (3) at two review calls, one CCR plus one cross-model review matches more planted errors than two CCR reviews (56.7% vs. 42.7%; Holm-adjusted p=.006), but not significantly more than two reviews by the top-tier cross-model reviewer, so model difference and reviewer capability are not separated. A lightweight cross-model reviewer scores no higher than same-model review. Withholding requirements from the reviewer raises F1 for the two lower tiers but not the top tier, in untested point estimates whose pattern depends on how failed sessions are scored. Before analysis we audited all session records, excluding one baseline run of uncertain provenance and 14 failed calls; results with all sessions are also reported. A partial check on public detector outputs from another benchmark neither replicates nor contradicts the main comparison. Records, artifacts, and scripts are available from the author on request.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-02 in cs.AI).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI); Software Engineering (cs.SE).
Pagina del lavoro: https://arxiv.org/abs/2610.01471 — 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 2026-10-06T10:21:28.186Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).