MedMax: Mixed-Modal Instruction Tuning for Training Biomedical Assistants

TL;DR

MedMax introduces a large-scale, diverse multimodal dataset for biomedical instruction tuning, significantly enhancing the performance of foundation models in biomedical visual question answering and related tasks.

Contribution

We created MedMax, a comprehensive multimodal biomedical dataset, and demonstrated its effectiveness in fine-tuning models for improved biomedical AI assistance.

Findings

26% performance improvement over Chameleon

18.3% improvement over GPT-4o in biomedical VQA

Diverse tasks across biomedical domains

Abstract

Recent advancements in mixed-modal generative have opened new avenues for developing unified biomedical assistants capable of analyzing biomedical images, answering complex questions about them, and generating multimodal patient reports. However, existing datasets face challenges such as small sizes, limited coverage of biomedical tasks and domains, and a reliance on narrow sources. To address these gaps, we present MedMax, a large-scale multimodal biomedical instruction-tuning dataset for mixed-modal foundation models. With 1.47 million instances, MedMax encompasses a diverse range of tasks, including interleaved image-text generation, biomedical image captioning and generation, visual chat, and report understanding. These tasks span knowledge across diverse biomedical domains, including radiology and histopathology, grounded in medical papers and YouTube videos. Subsequently, we fine-tune a mixed-modal foundation model on the MedMax dataset, achieving significant performance improvements: a 26% gain over the Chameleon model and an 18.3% improvement over GPT-4o across 12 downstream biomedical visual question-answering tasks. Finally, we introduce a unified evaluation suite for biomedical tasks to guide the development of mixed-modal biomedical AI assistants. The data, model, and code is available at https://mint-medmax.github.io/.