Go to ACL ARR 2025 May homepageHealthCards: Exploring Text-to-Image Generation as Visual Aids for Healthcare Knowledge Democratizing and Education
Abstract: The evolution of text-to-image (T2I) generation techniques has brought new capability for information visualization, and this advancement could have the potential to boost knowledge democratization and educational equity. In this paper, we envision these technologies as powerful tools to promote accessible healthcare knowledge education, which could not only serve the public but also more beneficial for communities in underserved regions and people with specific disabilities such as reading ability and attention limitations. We first explore how to harness recent T2I models to generate health knowledge flashcards, which are educational aids that aggregate knowledge with visually appealing and concise presentations in an image. Then, we curated a diverse and high quality healthcare knowledge flashcards datasets with 2034 samples from credible knowledge resources. We also validate the effectiveness of fine-tuning open-sourced models with our dataset to serve as a promising health flashcards generator. Our code is available at Anonymous Github: https://anonymous.4open.science/r/HealthCards
Paper Type: Long
Research Area: Multimodality and Language Grounding to Vision, Robotics and Beyond
Research Area Keywords: Healthcare Education, Text-to-Image generation, Knowledge Democratizing
Contribution Types: Publicly available software and/or pre-trained models, Data resources
Languages Studied: English
Submission Number: 7706
Loading