Jérôme Dockès

NeuroConText improves brain meta-analysis by using contrastive learning and large language models to better retrieve and reconstruct brain activation maps from neuroscience texts.

The paper provides tools for collecting, processing, and annotating biomedical literature to streamline text-mining and meta-science projects.

This paper reviews neuroimaging meta-analyses, finding that most follow PRISMA guidelines but have issues with pre-registration, search strategies, and study numbers.

This paper develops and evaluates methods for decoding functional connectivity gradients in fMRI data using meta-analytic techniques and k-means segmentation.