Skip to main content
Logo
The European High Performance Computing Joint Undertaking (EuroHPC JU)

Aloe Vera

50,000
Awarded Resources (in node hours)
Leonardo Booster
System Partition
3 June 2024 - 2 June 2025
Allocation Period

AI Technology: Natural Language Processing, Generative Language Modeling, Vision (image recognition, image generation, text recognition OCR, etc.)

 

This project aims to train a Medical-Vision Language Model (Med-VLM) that can process medical images and provide high-quality textual answers to medical questions in various languages.

The model will be open, free-to-use, reliable, and ethical. Unlike existing models like LLaVA-Med, Med-Flamingo, and Med-PaLM M, our Med-VLM will have localization skills for both inputs and outputs, making it more versatile and interpretable. 

The project aim is to improve healthcare accessibility, especially in areas with limited access to clinical experts, and to empower a new generation of open source tools. In addition to assessing its performance on medical datasets, the model will also be evaluated on benchmarks for biases and toxicity in language processing. 

This comprehensive evaluation ensures that the model not only achieves high accuracy in medical tasks but also mitigates biases and promotes ethical language generation, enhancing its applicability in diverse healthcare settings.