21 citations · 52 across the 15 of their papers we have counts for
17 papers
Learn from your own latents and not from tokens: A sample-complexity theory
Daniel J. Korchinski, Alessandro Favero, Matthieu Wyart
Generative models, from diffusion models to large language models, achieve remarkable performance but at a cost in training data orders of magnitude larger than what biological lea…
Task Addition and Weight Disentanglement in Closed-Vocabulary Models
Adam Hazimeh, Alessandro Favero, Pascal Frossard
Task arithmetic has recently emerged as a promising method for editing pre-trained \textit{open-vocabulary} models, offering a cost-effective alternative to standard multi-task fin…
Backdoor Unlearning by Linear Task Decomposition
Amel Abdelraheem, Alessandro Favero, Gerome Bovet +1
Foundation models have revolutionized computer vision by enabling broad generalization across diverse tasks. Yet, they remain highly susceptible to adversarial perturbations and ta…
The Physics of Data and Tasks: Theories of Locality and Compositionality in Deep Learning
Alessandro Favero
Deep neural networks have achieved remarkable success, yet our understanding of how they learn remains limited. These models can learn high-dimensional tasks, which is generally st…
MEMOIR: Lifelong Model Editing with Minimal Overwrite and Informed Retention for LLMs
Ke Wang, Yiming Qin, Nikolaos Dimitriadis +2
Language models deployed in real-world systems often require post-hoc updates to incorporate new or corrected knowledge. However, editing such models efficiently and reliably-witho…
Scaling Laws and Representation Learning in Simple Hierarchical Languages: Transformers vs. Convolutional Architectures
Francesco Cagnetta, Alessandro Favero, Antonio Sclocchi +1
How do neural language models acquire a language's structure when trained for next-token prediction? We address this question by deriving theoretical scaling laws for neural networ…
Bigger Isn't Always Memorizing: Early Stopping Overparameterized Diffusion Models
Alessandro Favero, Antonio Sclocchi, Matthieu Wyart
Diffusion probabilistic models have become a cornerstone of modern generative AI, yet the mechanisms underlying their generalization remain poorly understood. In fact, if these mod…
How Compositional Generalization and Creativity Improve as Diffusion Models are Trained
Alessandro Favero, Antonio Sclocchi, Francesco Cagnetta +2
Natural data is often organized as a hierarchical composition of features. How many samples do generative models need in order to learn the composition rules, so as to produce a co…
LiNeS: Post-training Layer Scaling Prevents Forgetting and Enhances Model Merging
Ke Wang, Nikolaos Dimitriadis, Alessandro Favero +3
Fine-tuning pre-trained models has become the standard approach to endow them with specialized knowledge, but it poses fundamental challenges. In particular, \textit{(i)} fine-tuni…
Probing the Latent Hierarchical Structure of Data via Diffusion Models
Antonio Sclocchi, Alessandro Favero, Noam Itzhak Levi +1
High-dimensional data must be highly structured to be learnable. Although the compositional and hierarchical nature of data is often put forward to explain learnability, quantitati…
Multi-Modal Hallucination Control by Visual Information Grounding
Alessandro Favero, Luca Zancato, Matthew Trager +5
Generative Vision-Language Models (VLMs) are prone to generate plausible-sounding textual answers that, however, are not always grounded in the input image. We investigate this phe…
A Phase Transition in Diffusion Models Reveals the Hierarchical Nature of Data
Antonio Sclocchi, Alessandro Favero, Matthieu Wyart
Understanding the structure of real data is paramount in advancing modern deep-learning methodologies. Natural data such as images are believed to be composed of features organized…
How Deep Neural Networks Learn Compositional Data: The Random Hierarchy Model
Francesco Cagnetta, Leonardo Petrini, Umberto M. Tomasini +2
Deep learning algorithms demonstrate a surprising ability to learn high-dimensional tasks from limited examples. This is commonly attributed to the depth of neural networks, enabli…
Task Arithmetic in the Tangent Space: Improved Editing of Pre-Trained Models
Guillermo Ortiz-Jimenez, Alessandro Favero, Pascal Frossard
Task arithmetic has recently emerged as a cost-effective and scalable approach to edit pre-trained models directly in weight space: By adding the fine-tuned weights of different ta…
What Can Be Learnt With Wide Convolutional Neural Networks?
Francesco Cagnetta, Alessandro Favero, Matthieu Wyart
Understanding how convolutional neural networks (CNNs) can efficiently learn high-dimensional functions remains a fundamental challenge. A popular belief is that these models harne…
Locality defeats the curse of dimensionality in convolutional teacher-student scenarios
Alessandro Favero, Francesco Cagnetta, Matthieu Wyart
Convolutional neural networks perform a local and translationally-invariant treatment of the data: quantifying which of these two aspects is central to their success remains a chal…
Relative stability toward diffeomorphisms indicates performance in deep nets
Leonardo Petrini, Alessandro Favero, Mario Geiger +1
Understanding why deep nets can classify data in large dimensions remains a challenge. It has been proposed that they do so by becoming stable to diffeomorphisms, yet existing empi…