Zobrazeno 1 - 10
of 156
pro vyhledávání: '"Claveau, Vincent"'
Distinctive Image Captioning: Leveraging Ground Truth Captions in CLIP Guided Reinforcement Learning
Training image captioning models using teacher forcing results in very generic samples, whereas more distinctive captions can be very useful in retrieval applications or to produce alternative texts describing images for accessibility. Reinforcement
Externí odkaz:
http://arxiv.org/abs/2402.13936
We present a hybrid approach to the automated measurement of vagueness and subjectivity in texts. We first introduce the expert system VAGO, we illustrate it on a small benchmark of fact vs. opinion sentences, and then test it on the larger French pr
Externí odkaz:
http://arxiv.org/abs/2309.06132
Autor:
Chaffin, Antoine, Scialom, Thomas, Lamprier, Sylvain, Staiano, Jacopo, Piwowarski, Benjamin, Kijak, Ewa, Claveau, Vincent
Language models generate texts by successively predicting probability distributions for next tokens given past ones. A growing field of interest tries to leverage external information in the decoding process so that the generated texts have desired p
Externí odkaz:
http://arxiv.org/abs/2204.11586
Autor:
Lamprier, Sylvain, Scialom, Thomas, Chaffin, Antoine, Claveau, Vincent, Kijak, Ewa, Staiano, Jacopo, Piwowarski, Benjamin
Generative Adversarial Networks (GANs) have known a tremendous success for many continuous generation tasks, especially in the field of image generation. However, for discrete outputs such as language, optimizing GANs remains an open problem with man
Externí odkaz:
http://arxiv.org/abs/2201.12320
The quality of artificially generated texts has considerably improved with the advent of transformers. The question of using these models to generate learning data for supervised learning tasks naturally arises. In this article, this question is expl
Externí odkaz:
http://arxiv.org/abs/2110.13016
Large language models (LM) based on Transformers allow to generate plausible long texts. In this paper, we explore how this generation can be further controlled at decoding time to satisfy certain constraints (e.g. being non-toxic, conveying certain
Externí odkaz:
http://arxiv.org/abs/2109.13582
Autor:
Claveau, Vincent
A well-known way to improve the performance of document retrieval is to expand the user's query. Several approaches have been proposed in the literature, and some of them are considered as yielding state-of-the-art results in IR. In this paper, we ex
Externí odkaz:
http://arxiv.org/abs/2012.08787
Autor:
Claveau, Vincent
De nombreuses applications du traitement automatique des langues (recherche d'information, traduction automatique, etc.) requièrent des ressources sémantiques spécifiques à leur tâche et à leur domaine. Pour répondre à ces besoins spécifique
Externí odkaz:
http://tel.archives-ouvertes.fr/tel-00524646
http://tel.archives-ouvertes.fr/docs/00/52/46/46/PDF/these_Vincent_Claveau.pdf
http://tel.archives-ouvertes.fr/docs/00/52/46/46/PDF/these_Vincent_Claveau.pdf
Publikováno v:
Conférence sur le Traitement Automatique des Langues Naturelles (TALN 2023)
Conférence sur le Traitement Automatique des Langues Naturelles (TALN 2023), Jun 2023, Paris, France
Conférence sur le Traitement Automatique des Langues Naturelles (TALN 2023), Jun 2023, Paris, France
International audience; The VAGO tool is an expert system for lexical vagueness detection that also measures the degree of subjectivity of the speech, as well as its level of detail. In this paper, we build a neural clone of VAGO, based on a BERT-lik
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::3a3e1ff024d07521eab8d84cbd610472
https://hal.science/hal-04109155/file/Icard&al_TALN2023.pdf
https://hal.science/hal-04109155/file/Icard&al_TALN2023.pdf
Autor:
Claveau, Vincent, Lefèvre, Sébastien
Publikováno v:
In Computer Speech & Language January 2015 29(1):63-80
