Multimodal foundation models serve numerous applications at the intersection of vision and
language. Still, despite being pretrained on extensive data, they become outdated over time …

Abstract Concept Bottleneck Models (CBMs) ground image classification on human-
understandable concepts to allow for interpretable model decisions as well as human …

Pretrained models have become a commodity and offer strong results on a broad range of
tasks. In this work, we focus on classification and seek to learn a unique encoder able to …

Knowledge distillation (KD) is the de facto standard for compressing large-scale models into
smaller ones. Prior works have explored ever more complex KD strategies involving different …

Pretrained models have become a commodity and offer strong results on a broad range of
tasks. In this work, we focus on classification and seek to learn a unique encoder able to …

Nowadays, Edge Intelligence has seen unprecedented growth in most of our daily life
applications. Traditionally, most applications required significant efforts into data collection …

Model merging combines multiple expert models-finetuned from a base foundation model
on diverse tasks and domains-into a single, more capable model. However, most existing …

Parameter-efficient finetuning (PEFT) has become ubiquitous to adapt foundation models to
downstream task requirements while retaining their generalization ability. However, the …

Recent advancements in large language and vision models have demonstrated
extraordinary capabilities, driving researchers to train increasingly larger models in pursuit …