no code implementations • WS 2019 • Bruno Gaume, Lydia Mai Ho-Dac, Ludovic Tanguy, C{\'e}cile Fabre, B{\'e}n{\'e}dicte Pierrejean, Nabil Hathout, J{\'e}r{\^o}me Farinas, Julien Pinquier, Lola Danet, Patrice P{\'e}ran, Xavier De Boissezon, M{\'e}lanie Jucla
This paper presents the first results of a multidisciplinary project, the {``}Evolex{''} project, gathering researchers in Psycholinguistics, Neuropsychology, Computer Science, Natural Language Processing and Linguistics.
no code implementations • WS 2019 • B{\'e}n{\'e}dicte Pierrejean, Ludovic Tanguy
We know that word embeddings trained using neural-based methods (such as word2vec SGNS) are sensitive to stability problems and that across two models trained using the exact same set of parameters, the nearest neighbors of a word are likely to change.
no code implementations • SEMEVAL 2018 • B{\'e}n{\'e}dicte Pierrejean, Ludovic Tanguy
Neural word embeddings models (such as those built with word2vec) are known to have stability problems: when retraining a model with the exact same hyperparameters, words neighborhoods may change.
no code implementations • NAACL 2018 • B{\'e}n{\'e}dicte Pierrejean, Ludovic Tanguy
We propose a method to study the variation lying between different word embeddings models trained with different parameters.
no code implementations • JEPTALNRECITAL 2018 • B{\'e}n{\'e}dicte Pierrejean, Ludovic Tanguy
Les mod{\`e}les vectoriels de s{\'e}mantique distributionnelle (ou word embeddings), notamment ceux produits par les m{\'e}thodes neuronales, posent des questions de reproductibilit{\'e} et donnent des repr{\'e}sentations diff{\'e}rentes {\`a} chaque utilisation, m{\^e}me sans modifier leurs param{\`e}tres.