Search Results for author: Guillaume Gravier

Found 20 papers, 1 papers with code

Filtrage et régularisation pour améliorer la plausibilité des poids d’attention dans la tâche d’inférence en langue naturelle (Filtering and regularization to improve the plausibility of attention weights in NLI)

no code implementations JEP/TALN/RECITAL 2022 Duc Hau Nguyen, Guillaume Gravier, Pascale Sébillot

Nous étudions la plausibilité d’un mécanisme d’attention pour une tâche d’inférence de phrases (entailment), c’est-à-dire sa capacité à fournir une explication plausible pour un humain de la relation entre deux phrases.

Natural Language Inference

Active Learning for Interactive Relation Extraction in a French Newspaper’s Articles

no code implementations RANLP 2021 Cyrielle Mallart, Michel Le Nouy, Guillaume Gravier, Pascale Sébillot

Relation extraction is a subtask of natural langage processing that has seen many improvements in recent years, with the advent of complex pre-trained architectures.

Active Learning Relation +2

Relation, es-tu l\`a ? D\'etection de relations par LSTM pour am\'eliorer l'extraction de relations (Relation, are you there ? LSTM-based relation detection to improve knowledge extraction )

no code implementations JEPTALNRECITAL 2020 Cyrielle Mallart, Michel Le Nouy, Guillaume Gravier, Pascale S{\'e}billot

Notre mod{\`e}le s{'}appuie sur le plus court chemin de d{\'e}pendances entre deux entit{\'e}s, mod{\'e}lis{\'e} par un LSTM et combin{\'e} avec les types des entit{\'e}s. Sur la t{\^a}che de d{\'e}tection de relations, nous obtenons de meilleurs r{\'e}sultats qu{'}un mod{\`e}le {\'e}tat de l{'}art pour la classification de relations, avec une robustesse accrue aux relations in{\'e}dites.

Classification General Classification +1

Rethinking deep active learning: Using unlabeled data at model training

1 code implementation19 Nov 2019 Oriane Siméoni, Mateusz Budnik, Yannis Avrithis, Guillaume Gravier

By systematically evaluating on a number of popular acquisition strategies and datasets, we find that the use of unlabeled data during model training brings a surprising accuracy improvement in image classification, compared to the differences between acquisition strategies.

Active Learning Image Classification

AI in the media and creative industries

no code implementations10 May 2019 Giuseppe Amato, Malte Behrmann, Frédéric Bimbot, Baptiste Caramiaux, Fabrizio Falchi, Ander Garcia, Joost Geurts, Jaume Gibert, Guillaume Gravier, Hadmut Holken, Hartmut Koenitz, Sylvain Lefebvre, Antoine Liutkus, Fabien Lotte, Andrew Perkis, Rafael Redondo, Enrico Turrin, Thierry Vieville, Emmanuel Vincent

Thanks to the Big Data revolution and increasing computing capacities, Artificial Intelligence (AI) has made an impressive revival over the past few years and is now omnipresent in both research and industry.

Language-based Construction of Explorable News Graphs for Journalists

no code implementations WS 2017 R{\'e}mi Bois, Guillaume Gravier, Eric Jamet, Emmanuel Morin, Pascale S{\'e}billot, Maxime Robert

Faced with ever-growing news archives, media professionals are in need of advanced tools to explore the information surrounding specific events.

Entity Extraction using GAN

Generative Adversarial Networks for Multimodal Representation Learning in Video Hyperlinking

no code implementations15 May 2017 Vedran Vukotic, Christian Raymond, Guillaume Gravier

We show that GANs can be used for multimodal representation learning and that they provide multimodal representations that are superior to representations obtained with multimodal autoencoders.

Information Retrieval Representation Learning +1

One-Step Time-Dependent Future Video Frame Prediction with a Convolutional Encoder-Decoder Neural Network

no code implementations14 Feb 2017 Vedran Vukotić, Silvia-Laura Pintea, Christian Raymond, Guillaume Gravier, Jan van Gemert

There is an inherent need for autonomous cars, drones, and other robots to have a notion of how their environment behaves and to anticipate changes in the near future.

Optical Flow Estimation

\'Evaluation dune nouvelle structuration th\'ematique hi\'erarchique des textes dans un cadre de r\'esum\'e automatique et de d\'etection d'ancres au sein de vid\'eos (Evaluation of a novel hierarchical thematic structuring of texts in the framework of text summarization and anchor detection for video hyperlinking )

no code implementations JEPTALNRECITAL 2016 Anca Simon, Guillaume Gravier, Pascale S{\'e}billot

automatique et de d{\'e}tection d{'}ancres au sein de vid{\'e}os Anca Simon1 Guillaume Gravier2 Pascale S{\'e}billot3 (1) Universit{\'e} de Rennes 1, IRISA {\&} INRIA Rennes, Campus de Beaulieu, 35042 Rennes, France (2) CNRS, IRISA {\&} INRIA Rennes, Campus de Beaulieu, 35042 Rennes, France (3) INSA, IRISA {\&} INRIA Rennes, Campus de Beaulieu, 35042 Rennes, France anca. simon@irisa. fr, guillaume. gravier@irisa. fr, pascale. sebillot@irisa. fr R {\'E}SUM{\'E} Dans cet article, nous {\'e}valuons, {\`a} travers son int{\'e}r{\^e}t pour le r{\'e}sum{\'e} automatique et la d{\'e}tection d{'}ancres dans des vid{\'e}os, le potentiel d{'}une nouvelle structure th{\'e}matique extraite de donn{\'e}es textuelles, compos{\'e}e d{'}une hi{\'e}rarchie de fragments th{\'e}matiquement focalis{\'e}s. Cette structure est produite par un algorithme exploitant les distributions temporelles d{'}apparition des mots dans les textes en se fondant sur une analyse de salves lexicales.

Text Summarization

Vers une typologie de liens entre contenus journalistiques

no code implementations JEPTALNRECITAL 2015 Remi Bois, Guillaume Gravier, Emmanuel Morin, Pascale S{\'e}billot

Nous pr{\'e}sentons une typologie de liens pour un corpus multim{\'e}dia ancr{\'e} dans le domaine journalistique.

The ETAPE speech processing evaluation

no code implementations LREC 2014 Olivier Galibert, Jeremy Leixa, Gilles Adda, Khalid Choukri, Guillaume Gravier

The ETAPE evaluation is the third evaluation in automatic speech recognition and associated technologies in a series which started with ESTER.

Automatic Speech Recognition Automatic Speech Recognition (ASR) +4

Bridging the gap between speech technology and natural language processing: an evaluation toolbox for term discovery systems

no code implementations LREC 2014 Bogdan Ludusan, Maarten Versteegh, Aren Jansen, Guillaume Gravier, Xuan-Nga Cao, Mark Johnson, Emmanuel Dupoux

The unsupervised discovery of linguistic terms from either continuous phoneme transcriptions or from raw speech has seen an increasing interest in the past years both from a theoretical and a practical standpoint.

Language Acquisition

The ETAPE corpus for the evaluation of speech-based TV content processing in the French language

no code implementations LREC 2012 Guillaume Gravier, Gilles Adda, Niklas Paulsson, Matthieu Carr{\'e}, Aude Giraudel, Olivier Galibert

The paper presents a comprehensive overview of existing data for the evaluation of spoken content processing in a multimedia framework for the French language.

Speech Recognition

Cannot find the paper you are looking for? You can Submit a new open access paper.