BERTrade: Using Contextual Embeddings to Parse Old French

no code implementations • LREC 2022 • Loïc Grobol, Mathilde Regnault, Pedro Ortiz Suarez, Benoît Sagot, Laurent Romary, Benoit Crabbé

The successes of contextual word embeddings learned by training large-scale language models, while remarkable, have mostly occurred for languages where significant amounts of raw texts are available and where annotated data in downstream tasks have a relatively regular spelling.

Dependency Parsing POS +3

Challenges of language change and variation: towards an extended treebank of Medieval French

no code implementations • WS 2019 • Mathilde Regnault, Sophie Pr{\'e}vost, Eric Villemonte de la Clergerie

Adaptation d'une m\'etagrammaire du fran\ccais contemporain au fran\ccais m\'edi\'eval (Adapting an existing metagrammar for Contemporary French to Medieval French)

no code implementations • JEPTALNRECITAL 2019 • Mathilde Regnault

Le fran{\c{c}}ais m{\'e}di{\'e}val se caract{\'e}rise par une importante variabilit{\'e} langagi{\`e}re.

NER

Vers l'annotation par le jeu de corpus (plus) complexes : le cas de la langue de sp\'ecialit\'e (Towards (more) complex corpora annotation using a game with a purpose : the case of scientific language)

no code implementations • JEPTALNRECITAL 2017 • Kar{\"e}n Fort, Bruno Guillaume, Nicolas Lefebvre, Laura Ram{\'\i}rez, Mathilde Regnault, Mary Collins, Oksana Gavrilova, Tanti Kristanti

Nous avons pr{\'e}c{\'e}demment montr{\'e} qu{'}il est possible de faire produire des annotations syntaxiques de qualit{\'e} par des participants {\`a} un jeu ayant un but.