Metraipolis › archive › dc_6beb2c41e376
When VLMs Trust Context: Evaluating Scene Text Recognition under Misleading Context — arXiv:2609.34781
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.34781
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T21:02:56.907Z
- SHA-256 of the archived text
354deec9602b79d239bc14240cf0b191526f312a79d70426e527175fd6c6701f- Stable URL of this page
https://metraipolis.com/documento/dc_6beb2c41e376
How to cite
When VLMs Trust Context: Evaluating Scene Text Recognition under Misleading Context — arXiv:2609.34781 — arXiv (en), https://arxiv.org/abs/2609.34781 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6beb2c41e376 (Metraipolis archive dc_6beb2c41e376, sha256 354deec9602b79d239bc14240cf0b191526f312a79d70426e527175fd6c6701f)
BibTeX
@misc{dc6beb2c41e376,
title = {When VLMs Trust Context: Evaluating Scene Text Recognition under Misleading Context — arXiv:2609.34781},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.34781},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6beb2c41e376; sha256 354deec9602b79d239bc14240cf0b191526f312a79d70426e527175fd6c6701f},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.34781. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Vision-language models (VLMs) can read text in natural scenes, but their predictions may be influenced by the surrounding context. When the printed text conflicts with what the scene suggests, a model may return a more plausible word instead of the shown text. We introduce SceneFaith, a benchmark of 781 generated scene images for studying this behavior. Each output is classified as Literal, Canonical, or Other, separating faithful transcription from context-consistent rewriting and ordinary recognition errors. Across 15 models from seven families, all models show rewriting on clear images, with rates ranging from 8.45\% to 58.51\%. Controlled experiments further show that surrounding context matters: removing surrounding scene information reduces rewriting and improves literal accuracy, while changing the scene around the same text patch can also change model outputs. Moreover, weakening the target text with blur increases rewriting. These results show that reliable scene-text recognition requires VLMs to balance visual character evidence with contextual information, preserving clear text while using context mainly when the visual evidence is uncertain. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.AI). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.34781 — 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 2026-10-07T20:29:03.157Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).