Resources

11 results for "vision-language models"

Database Credentialed Access

RadVLM Instruction Dataset

Nicolas Deperrois, Hidetoshi Matsuo, Samuel Ruiperez-Campillo, et al.

We release the RadVLM instruction dataset, a large-scale resource used to train the RadVLM model on diverse radiology tasks. The dataset contains 1,115,021 image–instruction pairs spanning five task families: (i) report generation from frontal CXRs …
chest x-rays vision-language models medical ai
Published: Sept. 25, 2025. Version: 1.0.0