11 papers
TORINO: Token Reduction via Interpretable Concept Overlap in Vision-Language Models
Riccardo Renzulli, Gabriele Spadaro, Shruthi Gowda +2
Vision-Language Models (VLMs) have demonstrated impressive capabilities across different tasks, but their computational cost is dominated by the large number of visual tokens fed t…
Look But Don't Touch with Sparse Autoencoders for Unlearning in Diffusion Models
Enrico Cassano, Riccardo Renzulli, Rayyan Ahmed +2
Sparse autoencoders (SAEs) have recently been proposed as interpretable tools for concept-level manipulation, under the assumption that isolated features can serve as controllable…
SAEmnesia: Erasing Concepts in Diffusion Models with Supervised Sparse Autoencoders
Enrico Cassano, Riccardo Renzulli, Marco Nurisso +3
Concept unlearning in diffusion models is hampered by feature splitting, where concepts are distributed across many latent features, making their removal challenging and computatio…
MedSAE: Dissecting MedCLIP Representations with Sparse Autoencoders
Riccardo Renzulli, Colas Lepoutre, Enrico Cassano +1
Artificial intelligence in healthcare requires models that are accurate and interpretable. We advance mechanistic interpretability in medical vision by applying Medical Sparse Auto…
GoDe: Gaussians on Demand for Progressive Level of Detail and Scalable Compression
Francesco Di Sario, Riccardo Renzulli, Marco Grangetto +2
Recent progress in compressing explicit radiance field representations, particularly 3D Gaussian Splatting, has substantially reduced memory consumption while improving real-time r…
Cardiac Output Prediction from Echocardiograms: Self-Supervised Learning with Limited Data
Adson Duarte, Davide Vitturini, Emanuele Milillo +9
Cardiac Output (CO) is a key parameter in the diagnosis and management of cardiovascular diseases. However, its accurate measurement requires right-heart catheterization, an invasi…