Scholar
Tom Dupré la Tour
Google Scholar ID: LuzAM-4AAAAJ
OpenAI
Citations & Impact
Citations
911
H-index
12
i10-index
16
Publications
20
Co-authors
18
list available
Publications
Resume (English only)
Academic Achievements
- Published papers including 'Persona features control emergent misalignment' in 2025; Won PhD thesis award - 1st prize in Signal, Image and Vision in 2019; Won PhD student award - 1st prize at Université Paris-Saclay STIC doctoral school in 2018.
Research Experience
- Joined OpenAI in 2023; Postdoc at the Gallant Lab, UC Berkeley from 2019 to 2021; Core developer of scikit-learn between 2015 and 2022.
Education
- Ph.D. in 2018 from Telecom ParisTech, France, supervised by Alexandre Gramfort and Yves Grenier; Graduated from EPFL in 2015; Graduated from École polytechnique in 2013.
Background
- Research Interests: Interpretability of language models, AI safety; Professional Field: Developing machine learning and signal processing methods for interpreting human/animal brain recordings (electrocorticography, magnetoencephalography, functional magnetic resonance imaging, neuron spikes) and silicon brain recordings (large language model activations); Brief Introduction: A research scientist at OpenAI.
Miscellany
- Blog post: 'Dask-Distributed and Joblib'
