Home

I am a Maître de Conférences (~associate professor) at Sorbonne Université. I conduct my research in the MLIA group at ISIR. I am a board member of the ATALA, a French association that promotes and animates research in natural language processing.

My research focus on Natural Language Processing and Machine Learning. I am especially interested in inference algorithms for structured prediction.

I also give training in deep learning for Natural Language Processing for the private sector via CNRS formation entreprise.

News

• November 6th, 2023: Talk @ ISIR: “Graph-based semantic parsing and compositional generalization”

• October, 2024: Preprint of our EMNLP 2023 paper: Structural generalization in COGS: Supertagging is (almost) all you need (Alban Petit, Caio Corro, François Yvon)

• July, 2023: Two (unrelated) ANR grants on amortized optimization and LLMs were accepted! I will hire PhD students on these topics in September 2024, feel free to contact me for an internship or PhD on one of these topics.

• April 28th, 2023: I will give a seminar at ALMAnaCH/INRIA Paris, slide are available here

• February 15th, 2023: Preprint of our TACL paper: On graph-based reentrancy-free semantic parsing (Alban Petit, Caio Corro)

• February 14th, 2023 (11:30 am): Santiago Herrera and myself will give a talk titled “Les enjeux de l'extraction de règles et de descriptions grammaticales à partir de treebanks : présentation de deux méthodes et de leurs limites”
  Salle de conférence du LISN (Site Belvédère) - Campus Universitaire bât.507 - Rue du Belvédère 91405 - Orsay

• January 25th, 2023: Preprint of my EACL 2023 paper: On the inconsistency of separable losses for structured prediction (Caio Corro)

Students and postdocs

Current:

• Benno Uthayasooriyar, PhD student (sept. 2022 - …) co-supervised with Franck Vermet and Antoine Ly

• Alban Petit, PhD student (oct. 2020 - …) co-supervised with François Yvon

Former:

• Yara Khater, Master intern (2022 & 2023)

• Emile Chapuis, postdoc (jan. 2022 - dec. 2023) co-advised with Nadi Tomeh

• Lili Lu, Master intern (2022)

• Lynda Attouche, Undergrad intern (2021)

• Alban Petit, Master intern (2020), now PhD student

• Nicolas Devatine, Master intern (2020), now Phd student @ IRIT

Funded projects

As (co-)PI:

• Semi-Amor: Structured Prediction in NLP via Semi-amortized Methods for Lagrangian Relaxation: coordinated by Joseph Le Roux and myself, funded by Agence National de la Recherche (ANR)

• Real-time speech recognition in resource-limited computing environments: coordinated by Lucas Ondel and myself, funded by Direction Générale de l'Armement (DGA)

• Linguistically motivated neural architectures for natural language processing: coordinated by Nadi Tomeh and myself, funded by Réseau Francilien en Sciences Informatiques (RFSI)

Others:

• InExtenso: Intrinsic and Extrinsic evaluation of biases in large language models: coordinated by Karën Fort, funded by Agence National de la Recherche (ANR)

• Autogramm: coordinated by Sylvain Kahane, funded by Agence National de la Recherche (ANR)

Program Committee

Area Chair/Action Editor:

• ACL Rolling Review, Oct. 2021 - … (including: ACL 2022, NAACL 2022, …)

• EACL 2023, Conference of the European Chapter of the Association for Computational Linguistics

• TALN 2022, Conférence sur le Traitement Automatique des Langues Naturelles

• TALN 2021, Conférence sur le Traitement Automatique des Langues Naturelles

• NAACL 2021, Annual Conference of the North American Chapter of the Association for Computational Linguistics

• ACL 2020, Annual Meeting of the Association for Computational Linguistics

Reviewer:

• 2022: EMNLP, Neurips

• 2021: ACL, EMNLP

• 2020: COLING

• 2019: CoNLL, EMNLP, ACL, NAACL, JMLR

• 2018: EMNLP, CoNLL, RJC

• 2017: ECML PKDD (sub-reviewer)

Quick Presentation

I obtained a Ph.D. in 2018 from Université Paris 13 under the supervision of Joseph Le Roux and Adeline Nazarenko. We worked with Mathieu Lacroix on using combinatorial optimization techniques (Lagrangian relaxation) to solve efficiently problems related to syntactic parsing with mildly context-sensitive grammars (e.g. Lexicalized Tree Adjoining Grammars).

My Ph.D. was supported by the Labex EFL, a public grant overseen by the French National Research Agency (ANR) as part of the Investissements d’Avenir program (ANR-10-LABX-0083).

From November 2017 to August 2018, I was postdoctoral researcher working with Ivan Titov. I was formally employed by the University of Amsterdam but I spent my time there as a visitor at the University of Edinburgh (EdinburghNLP group).

From September 2019 to August 2023, I was an associate professor at Université Paris-Saclay where I was the head of the Artificial Intelligence master. I conducted my research in the Spoken Language Processing / Traitement du Langage Parlé group at LISN (ex LIMSI). I am currently on leave from this position.