Metraipolis › archive › dc_2430ae898656

Seeing Red, Thinking Bad: Color Bias in Vision Language Models — arXiv:2608.14286

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.14286
License (as archived)
CC0-1.0
Archived
2026-10-19T02:21:28.071Z
SHA-256 of the archived text
f1932404f6b1e096e039fa8ee4344ce3e6b9a372561a0e7e8555c09213cadc30
Stable URL of this page
https://metraipolis.com/documento/dc_2430ae898656

How to cite

Seeing Red, Thinking Bad: Color Bias in Vision Language Models — arXiv:2608.14286 — arXiv (en), https://arxiv.org/abs/2608.14286 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2430ae898656 (Metraipolis archive dc_2430ae898656, sha256 f1932404f6b1e096e039fa8ee4344ce3e6b9a372561a0e7e8555c09213cadc30)
BibTeX
@misc{dc2430ae898656,
  title = {Seeing Red, Thinking Bad: Color Bias in Vision Language Models — arXiv:2608.14286},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.14286},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2430ae898656; sha256 f1932404f6b1e096e039fa8ee4344ce3e6b9a372561a0e7e8555c09213cadc30},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.14286. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Vision language models (VLMs) are increasingly used in industrial decision-making systems, such as recruitment support and recommendation. This motivates careful analysis of how VLMs process visual and textual information. In this work, we study how VLMs interpret text rendered as an image, and investigate the influence of visual styling biases. To this end, we introduce Stealth Visual Prompts, which subtly change visual styling of text, such as color and contrast, while preserving semantic content. Using these prompts, we systematically control the visual styling of words in text and measure their impact on the analysis performed by VLMs. We further analyze how such visual perturbations affect the latent representations of the vision encoder. From our experiments, we observed that coloring positive words in green consistently shifts sentiment predictions toward a positive direction. As a result, VLMs often fail to properly account for negative words present in the text. Our analysis suggests that this behavior is correlated with changes in the latent representations of the vision encoder induced by color variations. In addition, we show that reducing text--background contrast increases reliance on visually salient cues and leads to more incorrect Visual Question Answering (VQA) outputs. These results suggest that the visual styling of rendered text can guide VLMs' interpretation in ways that diverge from human semantic understanding. Project page: this https URL

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-17 in cs.AI).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI); Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2608.14286 — 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 2026-10-19T02:02:02.930Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).