collaborators

5 papers

cs.LG2026

High-Dimensional Theory of LoRA Fine-Tuning in a Solvable Attention Model

O. Duranthon, F. Boncoraglio, L. Zdeborová

We develop a high-dimensional statistical theory of low-rank adaptation (LoRA) in attention models, capturing the interplay between pre-training and fine-tuning. We introduce a sol…

stat.ML2026

The Rules-and-Facts Model for Simultaneous Generalization and Memorization in Neural Networks

Gabriele Farné, Fabrizio Boncoraglio, Lenka Zdeborová

A key capability of modern neural networks is their capacity to simultaneously learn underlying rules and memorize specific facts or exceptions. Yet, theoretical understanding of t…

stat.ML2025

Inference in Spreading Processes with Neural-Network Priors

Davide Ghio, Fabrizio Boncoraglio, Lenka Zdeborová

Stochastic processes on graphs are a powerful tool for modelling complex dynamical systems such as epidemics. A recent line of work focused on the inference problem where one aims…

stat.ML2025

Single-Head Attention in High Dimensions: A Theory of Generalization, Weights Spectra, and Scaling Laws

Fabrizio Boncoraglio, Vittorio Erba, Emanuele Troiani +3

Trained attention layers exhibit striking and reproducible spectral structure of the weights, including low-rank collapse, bulk deformation, and isolated spectral outliers, yet the…

cs.LG2025

Bayes optimal learning of attention-indexed models

Fabrizio Boncoraglio, Emanuele Troiani, Vittorio Erba +1

We introduce the attention-indexed model (AIM), a theoretical framework for analyzing learning in deep attention layers. Inspired by multi-index models, AIM captures how token-leve…