Scholar
Ambroise Odonnat
Google Scholar ID: M_OS-3kAAAAJ
Ph.D. Student - Noah's Ark Lab & Inria
Citations & Impact
Citations
112
H-index
5
i10-index
3
Publications
12
Co-authors
34
list available
Publications
Resume (English only)
Academic Achievements
- Received ICML Oral Award, ICASSP Oral Award, and QBIN Best Flash Talk Award for his research; one of his recent articles was featured in Forbes; multiple publications accepted or preprinted, including 'Easing Optimization Paths: A Circuit Perspective' accepted at ICASSP 2025, 'Clustering Heads' as a preprint, and 'Large Language Models as Markov Chains' also featured in Forbes.
Research Experience
- Currently a Ph.D. student at Huawei Noah’s Ark Lab & Inria in Paris; presented research at leading institutions such as EPFL, ENS Ulm, and Criteo; contributed to open-source libraries.
Education
- Graduated from Ecole des Ponts ParisTech in 2023 and holds a master’s degree from ENS Paris-Saclay in Mathematics, Vision, and Machine Learning (MVA). Supervised by Romain Tavenard, Laetitia Chapel, and Ievgen Redko.
Background
- Interested in improving the core understanding of Transformers, particularly large language models, out-of-distribution generalization, Transformer training and fine-tuning, Vision Transformers, and Time Series forecasting.
Miscellany
- Enjoys working both with a few collaborators and as part of a larger team; maintains a research blog named logB.
