skip to content
Home

Faculty of Mathematics

 

Dr Alessandro Favero

Physics-AI Fellow

Publications

How Compositional Generalization and Creativity Improve as Diffusion Models are Trained
A Favero, A Sclocchi, F Cagnetta, P Frossard, M Wyart
(2025)
(link to publication)
A phase transition in diffusion models reveals the hierarchical nature of data.
A Sclocchi, A Favero, M Wyart
– Proc Natl Acad Sci U S A
(2025)
122,
e2408799121
(doi: 10.1073/pnas.2408799121)
Computational complexity of deep learning: fundamental limitations and empirical phenomena
B Barak, A Carrell, A Favero, W Li, L Stephan, A Zlokapa
– Journal of Statistical Mechanics: Theory and Experiment
(2024)
2024,
104008
(doi: 10.1088/1742-5468/ad3a5b)
LiNeS: Post-training Layer Scaling Prevents Forgetting and Enhances Model Merging
K Wang, N Dimitriadis, A Favero, G Ortiz-Jimenez, F Fleuret, P Frossard
(2024)
(link to publication)
What can be learnt with wide convolutional neural networks?*
F Cagnetta, A Favero, M Wyart
– Journal of Statistical Mechanics: Theory and Experiment
(2024)
2024,
104020
(doi: 10.1088/1742-5468/ad65df)
Probing the Latent Hierarchical Structure of Data via Diffusion Models
A Sclocchi, A Favero, NI Levi, M Wyart
(2024)
(link to publication)
How Deep Neural Networks Learn Compositional Data: The Random Hierarchy Model
F Cagnetta, L Petrini, UM Tomasini, A Favero, M Wyart
– Physical Review X
(2024)
14,
031001
(doi: 10.1103/PhysRevX.14.031001)
Multi-Modal Hallucination Control by Visual Information Grounding
A Favero, L Zancato, M Trager, S Choudhary, P Perera, A Achille, A Swaminathan, S Soatto
– 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
(2024)
00,
14303
(doi: 10.1109/cvpr52733.2024.01356)
Multi-Modal Hallucination Control by Visual Information Grounding
A Favero, L Zancato, M Trager, S Choudhary, P Perera, A Achille, A Swaminathan, S Soatto
(2024)
(link to publication)
A Phase Transition in Diffusion Models Reveals the Hierarchical Nature of Data
A Sclocchi, A Favero, M Wyart
(2024)
(link to publication)
  • <
  • 2 of 3
  • >

Research Group

Relativity and Gravitation

Email

af940@cam.ac.uk

Room

B0.30

Personal homepage

https://alesfav.github.io/

Study at Cambridge

About the University

Research at Cambridge