SmartCaption AI - Enhancing Web Accessibility with Context-Aware Image Descriptions Using Large Language Models
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
The Internet provides vast amounts of information, services, and products. However, blind individuals and those with severe vision impairments face significant challenges in navi-gating web content, especially with understanding images. This paper introduces SmartCaption AI, an innovative solution that leverages Large Language Models (LLM) to generate descriptive text for images on web pages. By summarizing the content of a web page, SmartCaption AI provides relevant context for the LLM to produce accurate and meaningful image descriptions. These descriptions are seamlessly integrated into the web page's structure, allowing text- to-speech software to read them aloud to visually impaired users. SmartCaption AI offers several key contributions to web accessibility. It ensures the generated descriptions are contextually relevant, enhances the browsing experience by integrating real-time descriptions, and provides a universally accessible solution through a Chrome extension. This approach addresses the critical issue of missing or inadequate alternative text for images, thereby bridging the digital divide between sighted and visually impaired individuals. The results of our experiment demonstrated the effectiveness of SmartCaption AI, with an average score of 8.3/10, significantly outperforming state-of-art solutions: ImageToText (1.7/10) and AI-MCS (3.6/10). The source code of the tool is available on GitHub.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.003 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it