WebAccessVL: Violation-Aware VLM for Web Accessibility

TL;DR

This paper introduces WebAccessVL, a violation-aware vision-language model that automatically edits website HTML to fix accessibility violations while maintaining original design, significantly reducing violations and improving visual fidelity.

Contribution

WebAccessVL is a novel model that formulates website accessibility correction as a supervised, image-conditioned program synthesis task with an iterative checker-in-the-loop approach.

Findings

Achieves 96 ext% reduction in accessibility violations.

Outperforms GPT-5 by 87 ext% in violation reduction.

Perceptual study confirms better visual preservation.

Abstract

We present a vision-language model (VLM) that automatically edits website HTML to address violations of the Web Content Accessibility Guidelines 2 (WCAG2) while preserving the original design. We formulate this as a supervised image-conditioned program synthesis task, where the model learns to correct HTML given both the code and its visual rendering. We create WebAccessVL, a website dataset with manually corrected accessibility violations. We then propose a violation-conditioned VLM that further takes the detected violations' descriptions from a checker as input. This conditioning enables an iterative checker-in-the-loop refinement strategy at test time. We conduct extensive evaluation on both open API and open-weight models. Empirically, our method achieves 0.211 violations per website, a 96.0\% reduction from the 5.34 violations in raw data and 87\% better than GPT-5. A perceptual study also confirms that our edited websites better maintain the original visual appearance and content.