Language representations in deep learning algorithms and the brain

Autor: Caucheteux, Charlotte

Publikováno v: Artificial Intelligence [cs.AI]. Université Paris-Saclay, 2023. English. ⟨NNT : 2023UPASG031⟩

Recent deep language models -- like GPT-3 and ChatGPT -- are capable to produce text that closely resembles that of humans. Such similarity raises questions about how the brain and deep models process language, the mechanisms they use, and the intern

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=od______3515::2e4d2b3ffd1d164402035dad33c963cc
https://theses.hal.science/tel-04165471

Performance en classification de données textuelles des passages aux urgences des modèles BERT pour le français

Autor: Chenais, Gabrielle, Touchais, Hélène, Avalos, Marta, Bourdois, Loïck, Revel, Philippe, Gil-Jardiné, Cédric, Lagarde, Emmanuel

Publikováno v: PFIA 2021-Journée Santé et I.A.
PFIA 2021-Journée Santé et I.A., Journée organisée avec le soutien de l’Association française d’Informatique Médicale (AIM) et le Collège Science de l’Ingénierie des Connaissances de l’AFIA dans le cadre de la Plate-Forme Intelligence Artificielle (PFIA), Jun 2021, Bordeaux / Virtual, France

Contextualized language models based on the Transformer architecture such as BERT (Bidirectional Encoder Representations from Transformers) have achieved remarkable performances in various language processing tasks. CamemBERT and FlauBERT are pre-tra

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::71472c62262ebe42db370101540376ff
https://hal.inria.fr/hal-03276129

Performance of BERT models for French in the classification of textual data from emergency room visits

Autor: Chenais, Gabrielle, Touchais, Hélène, Avalos, Marta, Bourdois, Loïck, Revel, Philippe, Gil-Jardiné, Cédric, Lagarde, Emmanuel

Publikováno v: PFIA 2021-Journée Santé et I.A.
PFIA 2021-Journée Santé et I.A., Journée organisée avec le soutien de l’Association française d’Informatique Médicale (AIM) et le Collège Science de l’Ingénierie des Connaissances de l’AFIA dans le cadre de la Plate-Forme Intelligence Artificielle (PFIA), Jun 2021, Bordeaux / Virtual, France

Contextualized language models based on the Transformer architecture such as BERT (Bidirectional Encoder Representations from Transformers) have achieved remarkable performances in various language processing tasks. CamemBERT and FlauBERT are pre-tra

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=od______1398::71472c62262ebe42db370101540376ff
https://hal.inria.fr/hal-03276129

Restoring Arabic vowels through omission-tolerant dictionary lookup

Autor: Neme, Alexis Amid, Paumier, S��bastien

Publikováno v: Language Resources and Evaluation
Language Resources and Evaluation, Springer Verlag, 2020, 54, pp.487-551. ⟨10.1007/s10579-019-09464-6⟩

Vowels in Arabic are optional orthographic symbols written as diacritics above or below letters. In Arabic texts, typically more than 97 percent of written words do not explicitly show any of the vowels they contain; that is to say, depending on the

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::d7fbf28ac02bf96c8695061e623f390f
https://hal.archives-ouvertes.fr/hal-02113751

Développement d'un module de génération de paraphrases pour la data augmentation

Autor: Ratsiandavana, Sonia

Publikováno v: Sciences de l'Homme et Société. 2019

This paper summarizes the work completed during a six-month internship in the Research & Development team of Linagora. The aim of this project is to develop a paraphrasing tool able to expand the training datasets of the smart vocal assistant LinTO.

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=od______2592::800669a2f8a4c2abf6584ce811e99fb8
https://dumas.ccsd.cnrs.fr/dumas-02294883/document