Metraipolis › archive › dc_179e6cb2aced
CARLAR Influence: A Structured Hallucination Risk in Multimodal Cheque Extraction and an Exploratory Compressed-Image Countermeasure — arXiv:2610.27192
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.27192
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-21T07:07:31.736Z
- SHA-256 of the archived text
d105b5ebdf72c64868f480fa979bbbccdb87c4dcabd7fe9c4f18559e0ab8bc6c- Stable URL of this page
https://metraipolis.com/documento/dc_179e6cb2aced
How to cite
CARLAR Influence: A Structured Hallucination Risk in Multimodal Cheque Extraction and an Exploratory Compressed-Image Countermeasure — arXiv:2610.27192 — arXiv (en), https://arxiv.org/abs/2610.27192 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_179e6cb2aced (Metraipolis archive dc_179e6cb2aced, sha256 d105b5ebdf72c64868f480fa979bbbccdb87c4dcabd7fe9c4f18559e0ab8bc6c)
BibTeX
@misc{dc179e6cb2aced,
title = {CARLAR Influence: A Structured Hallucination Risk in Multimodal Cheque Extraction and an Exploratory Compressed-Image Countermeasure — arXiv:2610.27192},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.27192},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_179e6cb2aced; sha256 d105b5ebdf72c64868f480fa979bbbccdb87c4dcabd7fe9c4f18559e0ab8bc6c},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.27192. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Multimodal large language models (LLMs) are increasingly used for structured extraction from financial documents, but their behavior under visual ambiguity is poorly understood. We study CARLAR (Courtesy Amount Recognition and Legal Amount Recognition) influence: a model returns the same amount for both cheque fields even though their annotations differ, including cases with a blank annotation encoded as zero. Across ten visual prompting conditions, we evaluate GPT-4.1 and Claude 4.5 Sonnet on 243 primary cheques (124 with two observed unequal amounts) and 598 strict mismatches from two additional datasets. Under courtesy redaction, GPT-4.1 CARLAR reaches 73.7 percent (95 percent bootstrap interval: 68.3 to 79.0; p below 0.001 versus baseline). Model-reported confidence can remain high when predictions agree; expected calibration error reaches 0.93 among CARLAR cases. We also test a compressed-image few-shot intervention that presents examples as one visual table without model fine-tuning. On the 243-case cohort, Claude's Control condition increases paired joint extraction accuracy by 11.5 percentage points (95 percent interval: 6.2 to 16.9; adjusted p = 0.00054), but its strict-124 gain is smaller and not significant after correction. The intervention's effect on predicted-field agreement is model-dependent: it decreases for Claude and increases for GPT-4.1. These results motivate validation methods that account for visual context and dependencies between extracted fields. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.27192 — 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 2026-10-21T07:05:43.534Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).