no code implementations • 2 May 2024 • Théo Moutakanni, Piotr Bojanowski, Guillaume Chassagnon, Céline Hudelot, Armand Joulin, Yann Lecun, Matthew Muckley, Maxime Oquab, Marie-Pierre Revel, Maria Vakalopoulou
AI Foundation models are gaining traction in various applications, including medical fields like radiology.
no code implementations • 12 Apr 2024 • Arthur Ledaguenel, Céline Hudelot, Mostepha Khouadjia
Informed multi-label classification is a sub-field of neurosymbolic AI which studies how to leverage prior knowledge to improve neural classification systems.
no code implementations • 26 Mar 2024 • Eva Feillet, Adrian Popescu, Céline Hudelot
Our method outperforms competitive baselines, and performance is close to that of an oracle choosing the best algorithm in each setting.
2 code implementations • 20 Feb 2024 • Hippolyte Gisserot-Boukhlef, Manuel Faysse, Emmanuel Malherbe, Céline Hudelot, Pierre Colombo
Neural Information Retrieval (NIR) has significantly improved upon heuristic-based IR systems.
no code implementations • 20 Feb 2024 • Arthur Ledaguenel, Céline Hudelot, Mostepha Khouadjia
We develop a new multi-scale methodology to evaluate how the benefits of a neurosymbolic technique evolve with the scale of the network.
1 code implementation • 19 Feb 2024 • Nicolas Boizard, Kevin El Haddad, Céline Hudelot, Pierre Colombo
Deploying large language models (LLMs) of several billion parameters can be impractical in most industrial use cases due to constraints such as cost, latency limitations, and hardware accessibility.
1 code implementation • 1 Feb 2024 • Manuel Faysse, Patrick Fernandes, Nuno M. Guerreiro, António Loison, Duarte M. Alves, Caio Corro, Nicolas Boizard, João Alves, Ricardo Rei, Pedro H. Martins, Antoni Bigata Casademunt, François Yvon, André F. T. Martins, Gautier Viaud, Céline Hudelot, Pierre Colombo
We introduce CroissantLLM, a 1. 3B language model pretrained on a set of 3T English and French tokens, to bring to the research and industrial community a high-performance, fully open-sourced bilingual model that runs swiftly on consumer-grade local hardware.
1 code implementation • 21 Oct 2023 • Manuel Faysse, Gautier Viaud, Céline Hudelot, Pierre Colombo
Instruction Fine-Tuning (IFT) is a powerful paradigm that strengthens the zero-shot capabilities of Large Language Models (LLMs), but in doing so induces new evaluation metric requirements.
no code implementations • 22 Aug 2023 • Grégoire Petit, Michael Soumm, Eva Feillet, Adrian Popescu, Bertrand Delezoide, David Picard, Céline Hudelot
Our main finding is that the initial training strategy is the dominant factor influencing the average incremental accuracy, but that the choice of CIL algorithm is more important in preventing forgetting.
1 code implementation • CVPR 2023 • Malik Boudiaf, Etienne Bennequin, Myriam Tami, Antoine Toubhans, Pablo Piantanida, Céline Hudelot, Ismail Ben Ayed
We tackle the Few-Shot Open-Set Recognition (FSOSR) problem, i. e. classifying instances among a set of classes for which we only have a few labeled samples, while simultaneously detecting instances that do not belong to any known class.
no code implementations • 13 Sep 2022 • Thomas Cordier, Victor Bouvier, Gilles Hénaff, Céline Hudelot
Machine Learning models are prone to fail when test data are different from training data, a situation often encountered in real applications known as distribution shift.
1 code implementation • Findings (NAACL) 2022 • Jun Zhu, Céline Hudelot
Works on learning job title representation are mainly based on \textit{Job-Transition Graph}, built from the working history of talents.
no code implementations • 1 Feb 2022 • Umang Aggarwal, Adrian Popescu, Céline Hudelot
Here, we introduce a new active learning method which is designed for imbalanced datasets.
no code implementations • 1 Feb 2022 • Umang Aggarwal, Adrian Popescu, Eden Belouadah, Céline Hudelot
Since memory is bounded, old classes are learned with fewer images than new classes and an imbalance due to incremental learning is added to the initial dataset imbalance.
no code implementations • 18 Jan 2022 • Umang Aggarwal, Adrian Popescu, Céline Hudelot
It consists in learning a model on a small amount of annotated data (annotation budget) and in choosing the best set of points to annotate in order to improve the previous model and gain in generalization.
no code implementations • 9 Nov 2021 • Jun Zhu, Gautier Viaud, Céline Hudelot
The second module learns job seeker representations.
no code implementations • 15 Oct 2021 • Victor Pellegrain, Myriam Tami, Michel Batteux, Céline Hudelot
The increasing complexity of Industry 4. 0 systems brings new challenges regarding predictive maintenance tasks such as fault detection and diagnosis.
no code implementations • 30 Aug 2021 • Edoardo Ramalli, Alberto Parravicini, Guido Walter Di Donato, Mirko Salaris, Céline Hudelot, Marco Domenico Santambrogio
Drug repurposing is more relevant than ever due to drug development's rising costs and the need to respond to emerging diseases quickly.
no code implementations • 21 Jun 2021 • Martin Charachon, Paul-Henry Cournède, Céline Hudelot, Roberto Ardon
We show that visual explanation can be produced as the difference between two generated images obtained via two specific conditional generative models.
1 code implementation • 25 May 2021 • Etienne Bennequin, Victor Bouvier, Myriam Tami, Antoine Toubhans, Céline Hudelot
To classify query instances from novel classes encountered at test-time, they only require a support set composed of a few labelled samples.
1 code implementation • 26 Dec 2020 • Yassine Ouali, Céline Hudelot, Myriam Tami
In this paper, we explore contrastive learning for few-shot classification, in which we propose to use it as an additional auxiliary training objective acting as a data-dependent regularizer to promote more general and transferable features.
no code implementations • 21 Dec 2020 • Antoine Plumerault, Hervé Le Borgne, Céline Hudelot
Among the wide variety of image generative models, two models stand out: Variational Auto Encoders (VAE) and Generative Adversarial Networks (GAN).
no code implementations • 14 Dec 2020 • Martin Charachon, Céline Hudelot, Paul-Henry Cournède, Camille Ruppli, Roberto Ardon
From a given classifier, we train two generators to produce from an input image the so called similar and adversarial images.
no code implementations • 3 Dec 2020 • Victor Bouvier, Philippe Very, Clément Chastagnol, Myriam Tami, Céline Hudelot
First, we select for annotation target samples that are likely to improve the representations' transferability by measuring the variation, before and after annotation, of the transferability loss gradient.
1 code implementation • ECCV 2020 • Yassine Ouali, Céline Hudelot, Myriam Tami
In this work, we propose a new unsupervised image segmentation approach based on mutual information maximization between different constructed views of the inputs.
Ranked #5 on Unsupervised Semantic Segmentation on COCO-Stuff-3
no code implementations • 25 Jun 2020 • Yassine Ouali, Victor Bouvier, Myriam Tami, Céline Hudelot
Learning Invariant Representations has been successfully applied for reconciling a source and a target domain for Unsupervised Domain Adaptation.
no code implementations • 24 Jun 2020 • Victor Bouvier, Philippe Very, Clément Chastagnol, Myriam Tami, Céline Hudelot
The emergence of Domain Invariant Representations (IR) has improved drastically the transferability of representations from a labelled source domain to a new and unlabelled target domain.
1 code implementation • 9 Jun 2020 • Yassine Ouali, Céline Hudelot, Myriam Tami
Deep neural networks demonstrated their ability to provide remarkable performances on a wide range of supervised learning tasks (e. g., image classification) when trained on extensive collections of labeled data (e. g., ImageNet).
5 code implementations • CVPR 2020 • Yassine Ouali, Céline Hudelot, Myriam Tami
To leverage the unlabeled examples, we enforce a consistency between the main decoder predictions and those of the auxiliary decoders, taking as inputs different perturbed versions of the encoder's output, and consequently, improving the encoder's representations.
1 code implementation • ICLR 2020 • Antoine Plumerault, Hervé Le Borgne, Céline Hudelot
Recent deep generative models are able to provide photo-realistic images as well as visual or textual content embeddings useful to address various tasks of computer vision and natural language processing.
1 code implementation • 13 Jan 2020 • Béatrice Mazoyer, Nicolas Hervé, Céline Hudelot, Julia Cage
In this work, we evaluate the performance of recent text embeddings for the automatic detection of events in a stream of tweets.
no code implementations • 30 Dec 2019 • Régis Pierrard, Jean-Philippe Poli, Céline Hudelot
In this paper, we focus on organ annotation in medical images and we introduce a reasoning framework that is based on learning fuzzy relations on a small dataset for generating explanations.
no code implementations • 25 Sep 2019 • Victor Bouvier, Céline Hudelot, Clément Chastagnol, Philippe Very, Myriam Tami
Second, we show that learning weighted representations plays a key role in relaxing the constraint of invariance and then preserving the risk of compression.
no code implementations • 29 Jul 2019 • Victor Bouvier, Philippe Very, Céline Hudelot, Clément Chastagnol
Such approach consists in learning a representation of the data such that the label distribution conditioned on this representation is domain invariant.
no code implementations • 29 Jul 2019 • Victor Bouvier, Philippe Very, Céline Hudelot, Clément Chastagnol
Learning representations which remain invariant to a nuisance factor has a great interest in Domain Adaptation, Transfer Learning, and Fair Machine Learning.
no code implementations • 4 Oct 2018 • Julien Girard, Youssef Tamaazousti, Hervé Le Borgne, Céline Hudelot
This raises the question of how well the original representation is "universal", that is to say directly adapted to many different target-tasks.
1 code implementation • 27 Dec 2017 • Youssef Tamaazousti, Hervé Le Borgne, Céline Hudelot, Mohamed El Amine Seddik, Mohamed Tamaazousti
We also propose a unified framework of the methods based on the diversifying of the training problem.
no code implementations • 26 Feb 2015 • Marc Aiguier, Jamal Atif, Isabelle Bloch, Céline Hudelot
In this paper we address both the generalization of the well-known AGM postulates, and the definition of concrete and well-founded revision operators in different DL families.
no code implementations • 8 Feb 2015 • Marc Aiguier, Jamal Atif, Isabelle Bloch, Céline Hudelot
Belief revision of knowledge bases represented by a set of sentences in a given logic has been extensively studied but for specific logics, mainly propositional, and also recently Horn and description logics.
no code implementations • 18 Apr 2013 • Hichem Bannour, Céline Hudelot
Therefore, the built hierarchy is used in a semantic hierarchical classification framework for image annotation.