Search Results for author:
Found 28 papers, 6 papers with code
One Word, Two Sides: Traces of Stance in Contextualized Word Representations
The way we use words is influenced by our opinion.
Participation de l’équipe TGV à DEFT 2022 : Prédiction automatique de notes d’étudiants à des questionnaires en fonction du type de question (Team TGV at DEFT 2022 : automatic prediction of students’ grades according to the different question types)
Cet article présente l’approche de l’équipe TGV lors de sa participation à la tâche de base de DEFT 2022, dont l’objectif était de prédire automatiquement les notes obtenues par des étudiants sur la base de leurs réponses à des questionnaires.
Code-switched inspired losses for spoken dialog representations
Spoken dialogue systems need to be able to handle both multiple languages and multilinguality inside a conversation (e. g in case of code-switching).
Méta-apprentissage : classification de messages en catégories émotionnelles inconnues en entraînement (Meta-learning : Classifying Messages into Unseen Emotional Categories)
Dans cet article nous reproduisons un scénario d’apprentissage selon lequel les données cibles ne sont pas accessibles et seules des données connexes le sont.
Meta-learning for Classifying Previously Unseen Data Source into Previously Unseen Emotional Categories
In this paper, we place ourselves in a classification scenario in which the target classes and data type are not accessible during training.
EZCAT: an Easy Conversation Annotation Tool
We also present our first usage of EZCAT along with our annotation schema we used to annotate confidential customer service conversations.
Polysemy in Spoken Conversations and Written Texts
Our discourses are full of potential lexical ambiguities, due in part to the pervasive use of words having multiple senses.
The Impact of Word Splitting on the Semantic Content of Contextualized Word Representations
When deriving contextualized word representations from language models, a decision needs to be made on how to obtain one for out-of-vocabulary (OOV) words that are segmented into subwords.
End-to-end Supervised Prediction of Arbitrary-size Graphs with Partially-Masked Fused Gromov-Wasserstein Matching
We present a novel end-to-end deep learning-based approach for Supervised Graph Prediction (SGP).
Exploiting Edge Features in Graphs with Fused Network Gromov-Wasserstein Distance
Pairwise comparison of graphs is key to many applications in Machine learning ranging from clustering, kernel-based classification/regression and more recently supervised graph prediction.
A survey of neural models for the automatic analysis of conversation: Towards a better integration of the social sciences
In this paper we survey these neural architectures and what they have been applied to.
Few-Shot Emotion Recognition in Conversation with Sequential Prototypical Networks
We test this method on two datasets with different languages: daily conversations in English and customer service chat conversations in French.
Emotion Classification
Emotion Recognition in Conversation
+1
Improving Multimodal fusion via Mutual Dependency Maximisation
We demonstrate that our new penalties lead to a consistent improvement (up to $4. 3$ on accuracy) across a large variety of state-of-the-art models on two well-known sentiment analysis datasets: \texttt{CMU-MOSI} and \texttt{CMU-MOSEI}.
Code-switched inspired losses for generic spoken dialog representations
Spoken dialog systems need to be able to handle both multiple languages and multilinguality inside a conversation (\textit{e. g} in case of code-switching).
The importance of fillers for text representations of speech transcripts
While being an essential component of spoken language, fillers (e. g."um" or "uh") often remain overlooked in Spoken Language Understanding (SLU) tasks.
Hierarchical Pre-training for Sequence Labelling in Spoken Dialog
We obtain our representations with a hierarchical encoder based on transformer architectures, for which we extend two well-known pre-training objectives.
Ranked #1 on
Text Classification
on SILICONE Benchmark
Dialogue Act Classification
Emotion Recognition in Conversation
+1
Compositional Languages Emerge in a Neural Iterated Learning Model
The principle of compositionality, which enables natural language to represent complex concepts via a structured combination of simpler ones, allows us to convey an open-ended set of messages using a limited vocabulary.
Experimenting with Power Divergences for Language Modeling
In this paper, we experiment with several families (alpha, beta and gamma) of power divergences, generalized from the KL divergence, for learning language models with an objective different than standard MLE.
Learning with Noise-Contrastive Estimation: Easing training by learning to scale
Noise-Contrastive Estimation (NCE) is a learning criterion that is regularly used to train neural language models in place of Maximum Likelihood Estimation, since it avoids the computational bottleneck caused by the output softmax.
Algorithmes \`a base d'\'echantillonage pour l'entra\^\inement de mod\`eles de langue neuronaux (Here the title in English)
L{'}estimation contrastive bruit{\'e}e (NCE) et l{'}{\'e}chantillonage par importance (IS) sont des proc{\'e}dures d{'}entra{\^\i}nement bas{\'e}es sur l{'}{\'e}chantillonage, que l{'}on utilise habituellement {\`a} la place de l{'}estimation du maximum de vraisemblance (MLE) pour {\'e}viter le calcul du softmax lorsque l{'}on entra{\^\i}ne des mod{\`e}les de langue neuronaux.
Character and Subword-Based Word Representation for Neural Language Modeling Prediction
Most of neural language models use different kinds of embeddings for word prediction.
Adaptation au domaine pour l'analyse morpho-syntaxique (Domain Adaptation for PoS tagging)
Ce travail cherche {\`a} comprendre pourquoi les performances d{'}un analyseur morpho-syntaxiques chutent fortement lorsque celui-ci est utilis{\'e} sur des donn{\'e}es hors domaine.
Repr\'esentations continues d\'eriv\'ees des caract\`eres pour un mod\`ele de langue neuronal \`a vocabulaire ouvert (Opening the vocabulary of neural language models with character-level word representations)
Les repr{\'e}sentations continues des mots sont calcul{\'e}es {\`a} la vol{\'e}e {\`a} partir des caract{\`e}res les composant, gr{\`a}ce {\`a} une couche convolutionnelle suivie d{'}une couche de regroupement (pooling).
An experimental analysis of Noise-Contrastive Estimation: the noise distribution matters
Noise Contrastive Estimation (NCE) is a learning procedure that is regularly used to train neural language models, since it avoids the computational bottleneck caused by the output softmax.
