no code implementations • ACL (MetaNLP) 2021 • Oralie Cattan, Christophe Servan, Sophie Rosset
Supervised deep learning-based approaches have been applied to task-oriented dialog and have proven to be effective for limited domain and language applications when a sufficient number of training examples are available.
no code implementations • 19 Jul 2022 • Oralie Cattan, Sahar Ghannay, Christophe Servan, Sophie Rosset
In this paper, we propose a unified benchmark, focused on evaluating models quality and their ecological impact on two well-known French spoken language understanding tasks.
no code implementations • RANLP 2021 • Oralie Cattan, Christophe Servan, Sophie Rosset
In this paper, we establish a state-of-the-art of the efforts dedicated to the usability of Transformer-based models and propose to evaluate these improvements on the question-answering performances of French language which have few resources.
no code implementations • JEPTALNRECITAL 2020 • Oralie Cattan
{\'E}tendre les capacit{\'e}s d{'}adaptabilit{\'e} des syst{\`e}mes {\`a} toujours plus de nouveaux domaines sans donn{\'e}es de r{\'e}f{\'e}rence constitue une pierre d{'}achoppement de taille.
no code implementations • 6 Jul 2019 • Estelle Maudet, Oralie Cattan, Maureen de Seyssel, Christophe Servan
For this task, we propose an approach based on language models and evaluate the impact on the results of different preprocessings and matching techniques.
no code implementations • JEPTALNRECITAL 2019 • Estelle Maudet, Oralie Cattan, Maureen de Seyssel, Christophe Servan
Pour r{\'e}soudre cette t{\^a}che, nous proposons une approche reposant sur des mod{\`e}les de langue et {\'e}valuons l{'}impact de diff{\'e}rents pr{\'e}-traitements et de diff{\'e}rentes techniques d{'}appariement sur les r{\'e}sultats.