no code implementations • EMNLP (CODI) 2020 • Laurine Huber, Chaker Memmadi, Mathilde Dargnat, Yannick Toussaint
We introduce four tasks designed to determine which sentence encoders best capture discourse properties of sentences from scientific abstracts, namely coherence and cohesion between clauses of a sentence, and discourse relations within sentences.
no code implementations • LREC 2022 • Nyoman Juniarta, Olivier Bonami, Nabil Hathout, Fiammetta Namer, Yannick Toussaint
A family is regarded as a possible anomaly (meaning that it may have missing and/or spurious derivations) if its derivational graph is almost, but not completely identical to a large number of other families.
no code implementations • 7 Jun 2023 • Xavier Fontaine, Félix Gaschi, Parisa Rastin, Yannick Toussaint
However, other methods leveraging translation models can be used to perform NER without annotated data in the target language, by either translating the training set or test set.
1 code implementation • 5 Jun 2023 • Félix Gaschi, Patricio Cerda, Parisa Rastin, Yannick Toussaint
Namely, we find that realignment works better on tasks for which alignment is correlated with cross-lingual transfer when generalizing to a distant language and with smaller models, as well as when using a bilingual dictionary rather than FastAlign to extract realignment pairs.
1 code implementation • 19 Jul 2022 • Félix Gaschi, François Plesse, Parisa Rastin, Yannick Toussaint
Some Transformer-based models can perform cross-lingual transfer learning: those models can be trained on a specific task in one language and give relatively good results on the same task in another language, despite having been pre-trained on monolingual tasks only.
no code implementations • 24 Nov 2020 • Walid Hafiane, Joel Legrand, Yannick Toussaint, Adrien Coulet
Relation extraction (RE) consists in identifying and structuring automatically relations of interest from texts.
no code implementations • WS 2019 • Laurine Huber, Yannick Toussaint, Charlotte Roze, Mathilde Dargnat, Chlo{\'e} Braud
In this paper, we investigate similarities between discourse and argumentation structures by aligning subtrees in a corpus containing both annotations.
no code implementations • ICLR 2018 • Legrand Joël, Yannick Toussaint, Chedy Raïssi, Adrien Coulet
Indeed our approach leads to the best published performances for two biomedical RE tasks, and to state-of-the-art results for two other biomedical RE tasks, for which few annotated resources are available (less than 400 manually annotated sentences).
no code implementations • LREC 2016 • B{\'e}atrice Daille, Evelyne Jacquey, Ga{\"e}l Lejeune, Luis Felipe Melo, Yannick Toussaint
If a lexical unit is indeed a term of the domain, it is not true, even in a specialised corpus, that all its occurrences are terminological.