Search Results for author:
Found 49 papers, 14 papers with code
Identification de profil clinique du patient: Une approche de classification de séquences utilisant des modèles de langage français contextualisés (Identification of patient clinical profiles : A sequence classification approach using contextualised French language models )
Cet article présente un résumé de notre soumission pour Tâche 1 de DEFT 2021.
ON-TRAC’ systems for the IWSLT 2021 low-resource speech translation and multilingual speech translation shared tasks
This paper describes the ON-TRAC Consortium translation systems developed for two challenge tracks featured in the Evaluation Campaign of IWSLT 2021, low-resource speech translation and multilingual speech translation.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+3
Automatic Speech Recognition and Query By Example for Creole Languages Documentation
We investigate the exploitation of self-supervised models for two Creole languages with few resources: Gwadloupéyen and Morisien.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+1
WordNet and beyond: the case of lexical access
Next we will show under what conditions WN is suitable for word access, and finally we will present a roadmap showing the obstacles to be overcome to build a resource allowing the text producer to find the word s/he is looking for.
Visualizing Cross‐Lingual Discourse Relations in Multilingual TED Corpora
This paper presents an interactive data dashboard that provides users with an overview of the preservation of discourse relations among 28 language pairs.
LeBenchmark 2.0: a Standardized, Replicable and Enhanced Framework for Self-supervised Representations of French Speech
Self-supervised learning (SSL) is at the origin of unprecedented improvements in many different domains including computer vision and natural language processing.
UMLS-KGI-BERT: Data-Centric Knowledge Integration in Transformers for Biomedical Entity Recognition
In the biomedical domain, significant progress has been made in adapting this paradigm to NLP tasks that require the integration of domain-specific knowledge as well as statistical modelling of language.
Pre-training for Speech Translation: CTC Meets Optimal Transport
Nevertheless, CTC is only a partial solution and thus, in our second contribution, we propose a novel pre-training method combining CTC and optimal transport to further reduce this gap.
Effect Of Personalized Calibration On Gaze Estimation Using Deep-Learning
We trained a multi modal convolutional neural network and analysed its performance with and without calibration and this evaluation provides clear insights on how calibration improved the performance of the Deep Learning model in estimating gaze in the wild.
Lightweight Adapter Tuning for Multilingual Speech Translation
Adapter modules were recently introduced as an efficient alternative to fine-tuning in NLP.
Ranked #1 on
Speech-to-Text Translation
on MuST-C EN->ES
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+4
Do Multilingual Neural Machine Translation Models Contain Language Pair Specific Attention Heads?
Recent studies on the analysis of the multilingual representations focus on identifying whether there is an emergence of language-independent representations, or whether a multilingual model partitions its weights among different languages.
LeBenchmark: A Reproducible Framework for Assessing Self-Supervised Representation Learning from Speech
In this paper, we propose LeBenchmark: a reproducible framework for assessing SSL from speech.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+6
Dual-decoder Transformer for Joint Automatic Speech Recognition and Multilingual Speech Translation
We propose two variants of these architectures corresponding to two different levels of dependencies between the decoders, called the parallel and cross dual-decoder Transformers, respectively.
Ranked #1 on
Speech-to-Text Translation
on MuST-C EN->FR
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+3
FlauBERT : des mod\`eles de langue contextualis\'es pr\'e-entra\^\in\'es pour le fran\ccais (FlauBERT : Unsupervised Language Model Pre-training for French)
Les mod{\`e}les de langue pr{\'e}-entra{\^\i}n{\'e}s sont d{\'e}sormais indispensables pour obtenir des r{\'e}sultats {\`a} l{'}{\'e}tat-de-l{'}art dans de nombreuses t{\^a}ches du TALN.
Reconnaissance de parole beatbox\'ee \`a l'aide d'un syst\`eme HMM-GMM inspir\'e de la reconnaissance automatique de la parole (BEATBOX SOUNDS RECOGNITION USING A SPEECH-DEDICATED HMM-GMM BASED SYSTEM 1 Human beatboxing is a vocal art making use of speech organs to produce percussive sounds and imitate musical instruments)
Nous proposons un syst{\`e}me de reconnaissance des sons de beatbox s{'}inspirant de la reconnaissance automatique de la parole.
Providing Semantic Knowledge to a Set of Pictograms for People with Disabilities: a Set of Links between WordNet and Arasaac: Arasaac-WN
In order to make it possible to use pictograms automatically in NLP applications, we propose a database that links them to semantic knowledge.
Learning Term Discrimination
Document indexing is a key component for efficient information retrieval (IR).
FlauBERT: Unsupervised Language Model Pre-training for French
Language models have become a key step to achieve state-of-the art results in many different Natural Language Processing (NLP) tasks.
Ranked #1 on
Natural Language Inference
on XNLI French
WIKIR: A Python toolkit for building a large-scale Wikipedia-based English Information Retrieval Dataset
Since most standard ad-hoc information retrieval datasets publicly available for academic research (e. g. Robust04, ClueWeb09) have at most 250 annotated queries, the recent deep learning models for information retrieval perform poorly on these datasets.
The LIG system for the English-Czech Text Translation Task of IWSLT 2019
Therefore, we implemented a Transformer-based encoder-decoder neural system which is able to use the output of a pre-trained language model as input embeddings, and we compared its performance under three configurations: 1) without any pre-trained language model (constrained), 2) using a language model trained on the monolingual parts of the allowed English-Czech data (constrained), and 3) using a language model trained on a large quantity of external monolingual data (unconstrained).
ArbEngVec : Arabic-English Cross-Lingual Word Embedding Model
Word Embeddings (WE) are getting increasingly popular and widely applied in many Natural Language Processing (NLP) applications due to their effectiveness in capturing semantic properties of words; Machine Translation (MT), Information Retrieval (IR) and Information Extraction (IE) are among such areas.
Compression de vocabulaire de sens gr\^ace aux relations s\'emantiques pour la d\'esambigu\"\isation lexicale (Sense Vocabulary Compression through Semantic Knowledge for Word Sense Disambiguation)
En D{\'e}sambigu{\"\i}sation Lexicale (DL), les syst{\`e}mes supervis{\'e}s dominent largement les campagnes d{'}{\'e}valuation.
Apporter des connaissances s\'emantiques \`a un jeu de pictogrammes destin\'e \`a des personnes en situation de handicap : Un ensemble de liens entre Princeton WordNet et Arasaac, Arasaac-WN (Giving semantic knowledge to a set of pictograms for people with disabilities : a set of links between WordNet and Arasaac, Arasaac-WN )
Cet article pr{\'e}sente une ressource qui fait le lien entre WordNet et Arasaac, la plus grande base de pictogrammes librement disponible.
Sense Vocabulary Compression through the Semantic Knowledge of WordNet for Neural Word Sense Disambiguation
In this article, we tackle the issue of the limited quantity of manually sense annotated corpora for the task of word sense disambiguation, by exploiting the semantic relationships between senses such as synonymy, hypernymy and hyponymy, in order to compress the sense vocabulary of Princeton WordNet, and thus reduce the number of different sense tags that must be observed to disambiguate all words of the lexical database.
Ranked #1 on
Word Sense Disambiguation
on SemEval 2015 Task 13
Improving the Coverage and the Generalization Ability of Neural Word Sense Disambiguation through Hypernymy and Hyponymy Relationships
Our method leads to state of the art results on most WSD evaluation tasks, while improving the coverage of supervised systems, reducing the training time and the size of the models, without additional training data.
Ranked #2 on
Word Sense Disambiguation
on SemEval 2013 Task 12
Approche supervis\'ee \`a base de cellules LSTM bidirectionnelles pour la d\'esambigu\"\isation lexicale (LSTM Based Supervised Approach for Word Sense Disambiguation)
En d{\'e}sambigu{\"\i}sation lexicale, l{'}utilisation des r{\'e}seaux de neurones est encore peu pr{\'e}sente et tr{\`e}s r{\'e}cente.
Traduction automatique de corpus en anglais annot\'es en sens pour la d\'esambigu\"\isation lexicale d'une langue moins bien dot\'ee, l'exemple de l'arabe (Automatic Translation of English Sense Annotated Corpora for Word Sense Disambiguation of a Less Well-endowed Language, the Example of Arabic)
Nous {\'e}valuons la qualit{\'e} de nos syst{\`e}mes de d{\'e}sambigu{\"\i}sation gr{\^a}ce {\`a} un corpus d{'}{\'e}valuation en arabe nouvellement disponible.
Un corpus en arabe annot\'e manuellement avec des sens WordNet (Arabic Manually Sense Annotated Corpus with WordNet Senses)
OntoNotes comprend le seul corpus manuellement annot{\'e} en sens librement disponible pour l{'}arabe.
Système de traduction automatique statistique Anglais-Arabe
Machine translation (MT) is the process of translating text written in a source language into text in a target language.
LIM-LIG at SemEval-2017 Task1: Enhancing the Semantic Similarity for Arabic Sentences with Vectors Weighting
This article describes our proposed system named LIM-LIG.
Repr\'esentation vectorielle de sens pour la d\'esambigu\"\isation lexicale \`a base de connaissances (Sense Embeddings in Knowledge-Based Word Sense Disambiguation)
Dans cet article, nous proposons une nouvelle m{\'e}thode pour repr{\'e}senter sous forme vectorielle les sens d{'}un dictionnaire.
Uniformisation de corpus anglais annot\'es en sens (Unification of sense annotated English corpora for word sense disambiguation)
Pour la d{\'e}sambigu{\"\i}sation lexicale en anglais, on compte aujourd{'}hui une quinzaine de corpus annot{\'e}s en sens dans des formats souvent diff{\'e}rents et provenant de diff{\'e}rentes versions du Princeton WordNet.
Deep Investigation of Cross-Language Plagiarism Detection Methods
This paper is a deep investigation of cross-language plagiarism detection methods on a new recently introduced open dataset, which contains parallel and comparable collections of documents with multiple characteristics (different genres, languages and sizes of texts).
Comparison of Global Algorithms in Word Sense Disambiguation
We find that CSA, GA and SA all eventually converge to similar results (0. 98 F1 score), but CSA gets there faster (in fewer scorer calls) and reaches up to 0. 95 F1 before SA in fewer scorer calls.
CompiLIG at SemEval-2017 Task 1: Cross-Language Plagiarism Detection Methods for Semantic Textual Similarity
We present our submitted systems for Semantic Textual Similarity (STS) Track 4 at SemEval-2017.
Semantic Similarity of Arabic Sentences with Word Embeddings
Semantic textual similarity is the basis of countless applications and plays an important role in diverse areas, such as information retrieval, plagiarism detection, information extraction and machine translation.
Using Word Embedding for Cross-Language Plagiarism Detection
This paper proposes to use distributed representation of words (word embeddings) in cross-language textual similarity detection.
Am\'elioration de la traduction automatique d'un corpus annot\'e (Improvement of the automatic translation of an annotated corpus)
Dans cet article, nous pr{\'e}sentons une m{\'e}thode pour am{\'e}liorer la traduction automatique d{'}un corpus annot{\'e} et porter ses annotations de l{'}anglais vers une langue cible.
Extension lexicale de d\'efinitions gr\^ace \`a des corpus annot\'es en sens (Lexical Expansion of definitions based on sense-annotated corpus )
La proximit{\'e} s{\'e}mantique de deux d{\'e}finitions est {\'e}valu{\'e}e en comptant le nombre de mots communs dans les d{\'e}finitions correspondantes dans un dictionnaire.
A Multilingual, Multi-style and Multi-granularity Dataset for Cross-language Textual Similarity Detection
In this paper we describe our effort to create a dataset for the evaluation of cross-language textual similarity detection.
Cr\'eation rapide et efficace d'un syst\`eme de d\'esambigu\"\isation lexicale pour une langue peu dot\'ee
Nous pr{\'e}sentons une m{\'e}thode pour cr{\'e}er rapidement un syst{\`e}me de d{\'e}sambigu{\"\i}sation lexicale (DL) pour une langue L peu dot{\'e}e pourvu que l{'}on dispose d{'}un syst{\`e}me de traduction automatique statistique (TAS) d{'}une langue riche en corpus annot{\'e}s en sens (ici l{'}anglais) vers L. Il est, en effet, plus facile de disposer des ressources n{\'e}cessaires {\`a} la cr{\'e}ation d{'}un syst{\`e}me de TAS que des ressources d{\'e}di{\'e}es n{\'e}cessaires {\`a} la cr{\'e}ation d{'}un syst{\`e}me de DL pour la langue L. Notre m{\'e}thode consiste {\`a} traduire automatiquement un corpus annot{\'e} en sens vers la langue L, puis de cr{\'e}er le syst{\`e}me de d{\'e}sambigu{\"\i}sation pour L par des m{\'e}thodes supervis{\'e}es classiques.
