My Photo

William Léchelle

PhD student at Université de Montréal, advised by Philippe Langlais, at RALI.

Email:, or (last is lechelle)


Research interests

My research interests are in natural language understanding, in particular all forms of semantic parsing. Lately I've been working on Open Information Extraction.


Léchelle, W. and Langlais, P. (2018), Revisiting the Task of Scoring Open IE Relations.
Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018).
[pdf] [french slides] [poster]

CRM Industrial Problem Solving Workshop, working with the Co-operators.
[report - p27] [slides]

Thesis proposal (in french) : Pilotage empirique et regroupement d'information librement extraite.
[pdf] [slides]

Léchelle, W. and Langlais, P. (2016), An Informativeness Approach to Open IE Evaluation.
CICLing, Lecture Notes in Computer Science. Springer.
Best Verifiability, Reproducibility, and Working Description award, 3rd place.
[pdf] [slides] [bibtex] [code + data]

Léchelle, W., et Langlais, P. (2014), Utilisation de représentations de mots pour l’étiquetage de rôles sémantiques suivant FrameNet. (in french)
Actes de la 21e conférence sur le Traitement Automatique des Langues Naturelles, Marseille, France, Association pour le Traitement Automatique des Langues.
[pdf] [slides] [bibtex]


I've been a teaching assistant for the following :