Search Results for author:
Found 55 papers, 9 papers with code
Attention Modulation for Zero-Shot Cross-Domain Dialogue State Tracking
Dialog state tracking (DST) is a core step for task-oriented dialogue systems aiming to track the user’s current goal during a dialogue.
Analyzing BERT Cross-lingual Transfer Capabilities in Continual Sequence Labeling
Knowledge transfer between neural language models is a widely used technique that has proven to improve performance in a multitude of natural language tasks, in particular with the recent rise of large pre-trained language models like BERT.
Small Language Models are Good Too: An Empirical Study of Zero-Shot Classification
This study is part of the debate on the efficiency of large versus small language models for text classification by prompting. We assess the performance of small language models in zero-shot text classification, challenging the prevailing dominance of large models. Across 15 datasets, our investigation benchmarks language models from 77M to 40B parameters using different architectures and scoring functions.
New Semantic Task for the French Spoken Language Understanding MEDIA Benchmark
A combination ofmultiple datasets, including the MEDIA dataset, was suggested for training this joint model.
mALBERT: Is a Compact Multilingual BERT Model Still Worth It?
Within the current trend of Pretained Language Models (PLM), emerge more and more criticisms about the ethical andecological impact of such models.
On the Usability of Transformers-based models for a French Question-Answering task
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.
Benchmarking Transformers-based models on French Spoken Language Understanding tasks
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.
On the cross-lingual transferability of multilingual prototypical models across NLU tasks
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.
Overlap-aware low-latency online speaker diarization based on end-to-end local segmentation
We propose to address online speaker diarization as a combination of incremental clustering and local diarization applied to a rolling buffer updated every 500ms.
Evaluate On-the-job Learning Dialogue Systems and a Case Study for Natural Language Understanding
On-the-job learning consists in continuously learning while being used in production, in an open environment, meaning that the system has to deal on its own with situations and elements never seen before.
LIMSI\_UPV at SemEval-2020 Task 9: Recurrent Convolutional Neural Network for Code-mixed Sentiment Analysis
This paper describes the participation of LIMSI{\_}UPV team in SemEval-2020 Task 9: Sentiment Analysis for Code-Mixed Social Media Text.
Neural Networks approaches focused on French Spoken Language Understanding: application to the MEDIA Evaluation Task
In this paper, we present a study on a French Spoken Language Understanding (SLU) task: the MEDIA task.
LIMSI_UPV at SemEval-2020 Task 9: Recurrent Convolutional Neural Network for Code-mixed Sentiment Analysis
This paper describes the participation of LIMSI UPV team in SemEval-2020 Task 9: Sentiment Analysis for Code-Mixed Social Media Text.
A Metric Learning Approach to Misogyny Categorization
The task of automatic misogyny identification and categorization has not received as much attention as other natural language tasks have, even though it is crucial for identifying hate speech in social Internet interactions.
O\`u en sommes-nous dans la reconnaissance des entit\'es nomm\'ees structur\'ees \`a partir de la parole ? (Where are we in Named Entity Recognition from speech ?)
Les derni{\`e}res donn{\'e}es disponibles pour la REN structur{\'e}es {\`a} partir de la parole en fran{\c{c}}ais proviennent de la campagne d{'}{\'e}valuation ETAPE en 2012.
Where are we in Named Entity Recognition from Speech?
For this type of systems, we propose an original 3-pass approach.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+5
A Comparison of Metric Learning Loss Functions for End-To-End Speaker Verification
Despite the growing popularity of metric learning approaches, very little work has attempted to perform a fair comparison of these techniques for speaker verification.
DiaBLa: A Corpus of Bilingual Spontaneous Written Dialogues for Machine Translation
We provide a preliminary analysis of the corpus to confirm that the participants' judgments reveal perceptible differences in MT quality between the two MT systems used.
Survey on Evaluation Methods for Dialogue Systems
We cover each class by introducing the main technologies developed for the dialogue systems and then by presenting the evaluation methods regarding this class.
Natural language understanding for task oriented dialog in the biomedical domain in a low resources context
In the biomedical domain, the lack of sharable datasets often limit the possibility of developing natural language processing systems, especially dialogue applications and natural language understanding models.
Detecting context-dependent sentences in parallel corpora
In this article, we provide several approaches to the automatic identification of parallel sentences that require sentence-external linguistic context to be correctly translated.
\'Etiquetage en parties du discours de langues peu dot\'ees par sp\'ecialisation des plongements lexicaux (POS tagging for low-resource languages by adapting word embeddings )
Cet article pr{\'e}sente une nouvelle m{\'e}thode d{'}{\'e}tiquetage en parties du discours adapt{\'e}e aux langues peu dot{\'e}es : la d{\'e}finition du contexte utilis{\'e} pour construire les plongements lexicaux est adapt{\'e}e {\`a} la t{\^a}che, et de nouveaux vecteurs sont cr{\'e}{\'e}s pour les mots inconnus.
Automatic classification of doctor-patient questions for a virtual patient record query task
We present the work-in-progress of automating the classification of doctor-patient questions in the context of a simulated consultation with a virtual patient.
Apprendre des repr\'esentations jointes de mots et d'entit\'es pour la d\'esambigu\"\isation d'entit\'es (Combining Word and Entity Embeddings for Entity Linking)
La d{\'e}sambigu{\"\i}sation d{'}entit{\'e}s (ou liaison d{'}entit{\'e}s), qui consiste {\`a} relier des mentions d{'}entit{\'e}s d{'}un texte {\`a} des entit{\'e}s d{'}une base de connaissance, est un probl{\`e}me qui se pose, entre autre, pour le peuplement automatique de bases de connaissances {\`a} partir de textes.
Utterance Retrieval Based on Recurrent Surface Text Patterns
This paper investigates the use of recurrent surface text patterns to represent and index open-domain dialogue utterances for a retrieval system that can be embedded in a conversational agent.
\'Evaluation de l'apprentissage incr\'emental par analogie (Incremental Learning From Scratch Using Analogical Reasoning )
Cet article examine l{'}utilisation du raisonnement analogique dans le contexte de l{'}apprentissage incr{\'e}mental.
Un syst\`eme automatique de s\'election de r\'eponse en domaine ouvert int\'egrable \`a un syst\`eme de dialogue social (An automatic open-domain response selection system integrable to a social dialogue system)
Cette d{\'e}monstration pr{\'e}sente un syst{\`e}me de dialogue en domaine ouvert qui utilise une base d{'}exemples de dialogue automatiquement constitu{\'e}e depuis un corpus de sous-titres afin de g{\'e}rer un dialogue social de type « chatbot ».
Estimation de la qualit\'e d'un syst\`eme de reconnaissance de la parole pour une t\^ache de compr\'ehension (Quality estimation of a Speech Recognition System for a Spoken Language Understanding task)
Nous comparons ici diff{\'e}rentes m{\'e}triques, notamment le WER, NE-WER et ATENE m{\'e}trique propos{\'e}e r{\'e}cemment pour l{'}{\'e}valuation des syst{\`e}mes de reconnaissance de la parole {\'e}tant donn{\'e} une t{\^a}che de reconnaissance d{'}entit{\'e}s nomm{\'e}es.
Comparaison de listes d'erreurs de transcription automatique de la parole : quelle compl\'ementarit\'e entre les diff\'erentes m\'etriques ? (Comparing error lists for ASR systems : contribution of different metrics)
Le travail que nous pr{\'e}sentons ici s{'}inscrit dans le domaine de l{'}{\'e}valuation des syst{\`e}mes de reconnaissance automatique de la parole en vue de leur utilisation dans une t{\^a}che aval, ici la reconnaissance des entit{\'e}s nomm{\'e}es.
Generating Task-Pertinent sorted Error Lists for Speech Recognition
In the context of Automatic Speech Recognition (ASR) used as a first step towards Named Entity Recognition (NER) in speech, error seriousness is usually determined by their frequency, due to the use of the WER as metric to evaluate the ASR output, despite the emergence of more relevant measures in the literature.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+4
The CAMOMILE Collaborative Annotation Platform for Multi-modal, Multi-lingual and Multi-media Documents
In this paper, we describe the organization and the implementation of the CAMOMILE collaborative annotation framework for multimodal, multimedia, multilingual (3M) data.
Transfer-Based Learning-to-Rank Assessment of Medical Term Technicality
While measuring the readability of texts has been a long-standing research topic, assessing the technicality of terms has only been addressed more recently and mostly for the English language.
Managing Linguistic and Terminological Variation in a Medical Dialogue System
We introduce a dialogue task between a virtual patient and a doctor where the dialogue system, playing the patient part in a simulated consultation, must reconcile a specialized level, to understand what the doctor says, and a lay level, to output realistic patient-language utterances.
Purely Corpus-based Automatic Conversation Authoring
This system is used as a chatterbot system to collect a corpus of 41 open-domain textual dialogues with 27 human participants.
Named Entity Resources - Overview and Outlook
Recognition of real-world entities is crucial for most NLP applications.
Identification de facteurs de risque pour des patients diab\'etiques \`a partir de comptes-rendus cliniques par des approches hybrides
Dans cet article, nous pr{\'e}sentons les m{\'e}thodes que nous avons d{\'e}velopp{\'e}es pour analyser des comptes- rendus hospitaliers r{\'e}dig{\'e}s en anglais.
Un patient virtuel dialogant
Le d{\'e}monstrateur que nous d{\'e}crivons ici est un prototype de syst{\`e}me de dialogue dont l{'}objectif est de simuler un patient.
ETER : a new metric for the evaluation of hierarchical named entity recognition
We then introduce a new metric, the Entity Tree Error Rate (ETER), to evaluate hierarchical and structured named entity detection, classification and decomposition.
Human annotation of ASR error regions: Is ``gravity'' a sharable concept for human annotators?
This paper is concerned with human assessments of the severity of errors in ASR outputs.
Morpho-Syntactic Study of Errors from Speech Recognition System
The study provides an original standpoint of the speech transcription errors by focusing on the morpho-syntactic features of the erroneous chunks and of the surrounding left and right context.
The Quaero French Medical Corpus: A Ressource for Medical Entity Recognition and Normalization
Overall, a total of 26, 409 entity annotations were mapped to 5, 797 unique UMLS concepts.
Extended Named Entities Annotation on OCRed Documents: From Corpus Constitution to Evaluation Campaign
Within the framework of the Quaero project, we proposed a new definition of named entities, based upon an extension of the coverage of named entities as well as the structure of those named entities.
Named Entity Recognition (NER)
Optical Character Recognition (OCR)
Tree-Structured Named Entity Recognition on OCR Data: Analysis, Processing and Results
We evaluate our procedure for preprocessing OCR-ized data in two ways: in terms of perplexity and OOV rate of a language model on development and evaluation data, and in terms of the performance of the named entity detection system on the preprocessed data.
Designing French Tale Corpora for Entertaining Text To Speech Synthesis
This corpus is used for predicting expressive prosody in children tales, above the level of the sentence.
