Metraipolis › archive › dc_22e051de4bc2

What Breaks in Multi-Image Schema Application? Factorial Oracle Interventions Reveal an Instance-Information Bottleneck — arXiv:2610.19611

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.19611
License (as archived)
CC0-1.0
Archived
2026-10-15T21:56:16.408Z
SHA-256 of the archived text
49a575dd306415e6c61f1f6370cdb0391972b8fd37ab16f9fb6b258f2d41c138
Stable URL of this page
https://metraipolis.com/documento/dc_22e051de4bc2

How to cite

What Breaks in Multi-Image Schema Application? Factorial Oracle Interventions Reveal an Instance-Information Bottleneck — arXiv:2610.19611 — arXiv (en), https://arxiv.org/abs/2610.19611 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_22e051de4bc2 (Metraipolis archive dc_22e051de4bc2, sha256 49a575dd306415e6c61f1f6370cdb0391972b8fd37ab16f9fb6b258f2d41c138)
BibTeX
@misc{dc22e051de4bc2,
  title = {What Breaks in Multi-Image Schema Application? Factorial Oracle Interventions Reveal an Instance-Information Bottleneck — arXiv:2610.19611},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.19611},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_22e051de4bc2; sha256 49a575dd306415e6c61f1f6370cdb0391972b8fd37ab16f9fb6b258f2d41c138},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.19611. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Modern vision-language models can accept multi-image contexts, but aggregate accuracy rarely reveals where a multi-image reasoning pipeline fails. We study a controlled schema-application task in which a model must locate two target characters in a 16-image sequence, identify their visual attributes, consult an item-specific visual codebook, and compose two ordered nonce code names. We introduce a within-item 2-by-2 factorial Oracle design: Oracle-A supplies a verified textual schema, Oracle-B supplies a verified ordered target-appearance representation, and both interventions retain the original images and final question. Across 72 new items and 1,188 analyzed model responses, Oracle-B increased strict exact accuracy by 54.6 percentage points (95% confidence interval: 46.3 to 63.0) and outperformed Oracle-A by 60.6 points (Holm-adjusted p-value no greater than .0002). An auxiliary retrieval diagnostic reached only 49.1% ordered-appearance accuracy even after removing code-name generation. Providing both Oracle sources yielded 100% accuracy across all models. These results localize the dominant recoverable deficit to upstream ordered instance recovery rather than primarily to downstream schema use and composition.

Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2610.19611 — 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 2026-10-15T21:38:29.479Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).