Scholar
Nicolas Flammarion
Google Scholar ID: 7nVGYfgAAAAJ
EPFL
Machine Learning
Optimization
Statistics
Follow
Homepage
↗
Google Scholar
↗
Citations & Impact
All-time
Citations
6,328
H-index
33
i10-index
49
Publications
20
Co-authors
22
list available
Publications
27 items
Two-Timescale Fine-tuning Provably Learns New Features for Two-Layer ReLU Networks
2026
Cited
0
Statistical Benefits of Fine-Tuning from Pretrained Initialization in Diagonal Linear Networks
2026
Cited
0
Induction Heads Interpolate N-Grams
2026
Cited
0
Space-Efficient Language Generation in the Limit
2026
Cited
0
Learning What to Forget: Improving LLM Unlearning via Learned Token-Level Importance
2026
Cited
0
(How) Learning Rates Regulate Catastrophic Overtraining
2026
Cited
0
Transformers Learn Latent Mixture Models In-Context via Mirror Descent
2026
Cited
0
Gradient Flow Polarizes Softmax Outputs towards Low-Entropy Solutions
2026
Cited
0
Load more
Co-authors
20 total
Maksym Andriushchenko
ELLIS Institute Tübingen & Max Planck Institute for Intelligent Systems
Francesco Croce
EPFL
Matthias Hein
Professor of Computer Science, University of Tübingen
Francis Bach
Inria - Ecole Normale Supérieure
Peter Bartlett
Professor, EECS and Statistics, UC Berkeley
Michael I. Jordan
Professor of Electrical Engineering and Computer Sciences and Professor of Statistics, UC Berkeley
Scott Pesme
Post-Doc at Inria Grenoble
Aditya Varre
PhD student, EPFL
