Virginie Do

About

I am an AI Research Scientist at Meta working on LLM alignment. I am based in London, UK.

In 2023, I completed my PhD at PSL University (Université Paris Dauphine) and Meta AI (Facebook AI Research) in Paris, advised by Nicolas Usunier (Meta AI), Jamal Atif and Jérôme Lang (PSL University, CNRS).

My PhD focused on fairness in recommender systems and blends economic concepts with convex optimization and bandit techniques. An overview can be found in this blogpost.

Prior to my PhD, I received an MSc in Social Data Science from the University of Oxford, an MSc/BSc in Applied Mathematics from Ecole Polytechnique (Diplôme d'Ingénieur) , and a BA in Philosophy from Université Paris Nanterre.

Publications & preprints

Fairness in recommender systems: Insights from social choice. PhD thesis. [pdf]
Virginie Do.

Group fairness without demographics using social networks. ACM FAccT 2023. [pdf]
David Liu, Virginie Do, Nicolas Usunier and Maximilian Nickel.

Contextual bandits with concave rewards, and an application to fair ranking. ICLR 2023. [pdf]
Virginie Do Elvis Dohmatob, Matteo Pirotta, Alessandro Lazaric and Nicolas Usunier.

Optimizing generalized Gini indices for fairness in rankings. ACM SIGIR 2022. [pdf]
Virginie Do and Nicolas Usunier.

Fast online ranking with fairness of exposure. ACM FAccT 2022. [pdf]
Nicolas Usunier, Virginie Do, Elvis Dohmatob.

Online approval committee elections. IJCAI 2022. [pdf]
Virginie Do, Matthieu Hervouin, Jérôme Lang, and Piotr Skowron.

Online certification of preference-based fairness for personalized recommender systems. AAAI 2022. [pdf]
Virginie Do, Sam Corbett-Davies, Jamal Atif, and Nicolas Usunier.
🏆 Outstanding paper award (1 award / 1.3k accepted papers and 9.2k submissions).

Two-sided fairness in rankings via Lorenz dominance. NeurIPS 2021. [pdf]
Virginie Do, Sam Corbett-Davies, Jamal Atif, and Nicolas Usunier.

Online selection of diverse committees. IJCAI 2021. [pdf]
Virginie Do, Jamal Atif, Jérôme Lang, and Nicolas Usunier.

e-ViL: A Dataset and Benchmark for Natural Language Explanations in Vision-Language Tasks. ICCV 2021. [pdf]
Maxime Kayser, Oana-Maria Camburu, Leonard Salewski, Cornelius Emde, Virginie Do, Zeynep Akata, Thomas Lukasiewicz.

Service

I co-chaired the workshop Responsible Decision-Making in Dynamic Environments at ICML 2022 in Baltimore, organized with Thorsten Joachims, Alessandro Lazaric, Joelle Pineau, Matteo Pirotta, Harsh Satija and Nicolas Usunier.

Reviewer: NeurIPS 2021-2022, ICML 2022, IJCAI 2023, FAccT 2022-2023.