no code implementations • JEP/TALN/RECITAL 2021 • Philippe Suignard, Alexandra Benamar, Nazim Messous, Clément Christophe, Marie Jubault, Meryl Bothua
Ce papier présente la participation d’EDF R&D à la campagne d’évaluation DEFT 2021.
no code implementations • JEP/TALN/RECITAL 2022 • Alexandra Benamar, Cyril Grouin, Meryl Bothua, Anne Vilnat
Dans cet article, nous étudions les stéréotypes de genre qui existent dans des modèles Word2Vec.
1 code implementation • LREC 2022 • Alexandra Benamar, Cyril Grouin, Meryl Bothua, Anne Vilnat
Our experiments have led to exciting findings that showed: (1) It is easier to improve the representation of new words (A and B) than it is for words that already exist in the vocabulary of the Transformer models (C), (2) To ameliorate the representation of OOVs, the most effective method relies on adding external morpho-syntactic context rather than improving the semantic understanding of the words directly (fine-tuning) and (3) We cannot foresee the impact of minor misspellings in words because similar misspellings have different impacts on their representation.