Search Results for author:
Found 23 papers, 5 papers with code
Décodage guidé par un discriminateur avec le Monte Carlo Tree Search pour la génération de texte contrainte (Discriminator-guided decoding with Monte Carlo Tree Search for constrained text generation )
Dans cet article, nous explorons comment contrôler la génération de texte au moment du décodage pour satisfaire certaines contraintes (e. g. être non toxique, transmettre certaines émotions...), sans nécessiter de ré-entrainer le modèle de langue.
La génération de textes artificiels en substitution ou en complément de données d’apprentissage (Generating artificial texts as substitution or complement of training data )
(ii) peuvent-elles remplacer les données d’origines quand ces dernières ne peuvent pas être distribuées, par exemple pour des raisons de confidentialité ?
Choisir le bon co-équipier pour la génération coopérative de texte (Choosing The Right Teammate For Cooperative Text Generation)
Nous évaluons leurs avantages et inconvénients, en explorant leur précision respective sur des tâches de classification, ainsi que leur impact sur la génération coopérative et leur coût de calcul, dans le cadre d’une stratégie de décodage état de l’art, basée sur une recherche arborescente de Monte-Carlo (MCTS).
Distinctive Image Captioning: Leveraging Ground Truth Captions in CLIP Guided Reinforcement Learning
Secondly, they can serve as additional trajectories in the RL strategy, resulting in a teacher forcing loss weighted by the similarity of the GT to the image.
MAAIP: Multi-Agent Adversarial Interaction Priors for imitation from fighting demonstrations for physics-based characters
In this paper, we propose a novel Multi-Agent Generative Adversarial Imitation Learning based approach that generalizes the idea of motion imitation for one character to deal with both the interaction and the motions of the multiple physics-based characters.
Building Flyweight FLIM-based CNNs with Adaptive Decoding for Object Detection
State-of-the-art (SOTA) object detection methods have succeeded in several applications at the price of relying on heavyweight neural networks, which makes them inefficient and inviable for many applications with computational resource constraints.
Teach me how to Interpolate a Myriad of Embeddings
Finally, to address inconsistencies due to linear target interpolation, we introduce a self-distillation approach to generate and interpolate synthetic targets.
Which Discriminator for Cooperative Text Generation?
Language models generate texts by successively predicting probability distributions for next tokens given past ones.
Generative Cooperative Networks for Natural Language Generation
Generative Adversarial Networks (GANs) have known a tremendous success for many continuous generation tasks, especially in the field of image generation.
Generating artificial texts as substitution or complement of training data
The quality of artificially generated texts has considerably improved with the advent of transformers.
AlignMix: Improving representations by interpolating aligned features
Mixup is a powerful data augmentation method that interpolates between two or more examples in the input or feature space and between the corresponding target labels.
PPL-MCTS: Constrained Textual Generation Through Discriminator-Guided MCTS Decoding
without fine-tuning the LM.
It Takes Two to Tango: Mixup for Deep Metric Learning
In this work, we aim to bridge this gap and improve representations using mixup, which is a powerful data augmentation approach interpolating two or more examples and corresponding target labels at a time.
Ranked #8 on
Metric Learning
on CUB-200-2011
(using extra training data)
AlignMixup: Improving Representations By Interpolating Aligned Features
Mixup is a powerful data augmentation method that interpolates between two or more examples in the input or feature space and between the corresponding target labels.
Ranked #1 on
Representation Learning
on CIFAR10
Detecting Human-Object Interaction with Mixed Supervision
Human object interaction (HOI) detection is an important task in image understanding and reasoning.
Unsupervised part learning for visual recognition
This strategy opens the door to the use of PBM in new applications for which the notion of image categories is irrelevant, such as instance-based image retrieval, for example.
Direct vs. indirect evaluation of distributional thesauri
In this paper, we address the problem of the evaluation of such thesauri or embedding models and compare their results.
M\'edias traditionnels, m\'edias sociaux : caract\'eriser la r\'einformation (Traditional medias, social medias : characterizing reinformation)
Nous pr{\'e}sentons d{'}autre part quelques exp{\'e}riences de d{\'e}tection automatique des messages issus des m{\'e}dias de r{\'e}information, en {\'e}tudiant notamment l{'}influence d{'}attributs de surface et d{'}attributs portant plus sp{\'e}cifiquement sur le contenu de ces messages.
Distributional Thesauri for Information Retrieval and vice versa
In this paper, we address the problem of building and evaluating such thesauri with the help of Information Retrieval (IR) concepts.
Strat\'egies de s\'election des exemples pour l'apprentissage actif avec des champs al\'eatoires conditionnels
D{'}autre part, nous d{\'e}taillons une m{\'e}thode originale de s{\'e}lection s{'}appuyant sur un crit{\`e}re de respect des proportions dans les jeux de donn{\'e}es manipul{\'e}s. Le bien- fond{\'e} de ces propositions est v{\'e}rifi{\'e} au travers de plusieurs t{\^a}ches et jeux de donn{\'e}es, incluant reconnaissance d{'}entit{\'e}s nomm{\'e}es, chunking, phon{\'e}tisation, d{\'e}sambigu{\"\i}sation de sens.
Generating and using probabilistic morphological resources for the biomedical domain
In most Indo-European languages, many biomedical terms are rich morphological structures composed of several constituents mainly originating from Greek or Latin.
