Salvato in:
Dettagli Bibliografici
Autori principali: Renzulli, Riccardo, Lepoutre, Colas, Cassano, Enrico, Grangetto, Marco
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2510.26411
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916037475172352
author Renzulli, Riccardo
Lepoutre, Colas
Cassano, Enrico
Grangetto, Marco
author_facet Renzulli, Riccardo
Lepoutre, Colas
Cassano, Enrico
Grangetto, Marco
contents Artificial intelligence in healthcare requires models that are accurate and interpretable. We advance mechanistic interpretability in medical vision by applying Medical Sparse Autoencoders (MedSAEs) to the latent space of MedCLIP, a vision-language model trained on chest radiographs and reports. To quantify interpretability, we propose an evaluation framework that combines correlation metrics, entropy analyses, and automated neuron naming via the MedGemma foundation model. Experiments on the CheXpert dataset show that MedSAE neurons achieve higher monosemanticity and interpretability than raw MedCLIP features. Our findings bridge high-performing medical AI and transparency, offering a scalable step toward clinically reliable representations. The source code supporting the findings of this study is available at https://github.com/EIDOSLAB/MedSAE.
format Preprint
id arxiv_https___arxiv_org_abs_2510_26411
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MedSAE: Dissecting MedCLIP Representations with Sparse Autoencoders
Renzulli, Riccardo
Lepoutre, Colas
Cassano, Enrico
Grangetto, Marco
Artificial Intelligence
Artificial intelligence in healthcare requires models that are accurate and interpretable. We advance mechanistic interpretability in medical vision by applying Medical Sparse Autoencoders (MedSAEs) to the latent space of MedCLIP, a vision-language model trained on chest radiographs and reports. To quantify interpretability, we propose an evaluation framework that combines correlation metrics, entropy analyses, and automated neuron naming via the MedGemma foundation model. Experiments on the CheXpert dataset show that MedSAE neurons achieve higher monosemanticity and interpretability than raw MedCLIP features. Our findings bridge high-performing medical AI and transparency, offering a scalable step toward clinically reliable representations. The source code supporting the findings of this study is available at https://github.com/EIDOSLAB/MedSAE.
title MedSAE: Dissecting MedCLIP Representations with Sparse Autoencoders
topic Artificial Intelligence
url https://arxiv.org/abs/2510.26411