The Rediscovery Hypothesis: Language Models Need to Meet Linguistics

Autor: Vassilina Nikoulina, Maxat Tezekbayev, Nuradil Kozhakhmet, Madina Babazhanova, Matthias Gallé, Zhenisbek Assylbekov

There is an ongoing debate in the NLP community whether modern language models contain linguistic knowledge, recovered through so-called probes. In this paper, we study whether linguistic knowledge is a necessary condition for the good performance of

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::1d1e1ae8b72966cdf13e3e53f3b07acd
http://arxiv.org/abs/2103.01819

Monolingual Adapters for Zero-Shot Neural Machine Translation

Autor: Jerin Philip, Laurent Besacier, Alexandre Berard, Matthias Gallé

Publikováno v: Philip, J, Berard, A, Gallé, M & Besacier, L 2020, Monolingual Adapters for Zero-Shot Neural Machine Translation . in Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) . Online, pp. 4465-4470, The 2020 Conference on Empirical Methods in Natural Language Processing, Virtual conference, 16/11/20 . https://doi.org/10.18653/v1/2020.emnlp-main.361
EMNLP (Empirical Methods for Natural Language Processing)
EMNLP (Empirical Methods for Natural Language Processing), Nov 2020, Virtual, France
EMNLP (1)

International audience; We propose a novel adapter layer formalism for adapting multilingual models. They are more parameter-efficient than existing adapter layers while obtaining as good or better performance. The layers are specific to one language

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::42acc31213743e9ee9b05dbffd32168b
https://www.pure.ed.ac.uk/ws/files/192697520/Monolingual_Adapters_PHILIP_DOA14092020_VOR_CC_BY.pdf

Self-Supervised and Controlled Multi-Document Opinion Summarization

Autor: Maximin Coavoux, Matthias Gallé, Hady Elsahar, Jos Rozen

Publikováno v: Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume
Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume, Apr 2021, Online, Unknown Region. pp.1646--1662
EACL

We address the problem of unsupervised abstractive summarization of collections of user generated reviews with self-supervision and control. We propose a self-supervised setup that considers an individual document as a target summary for a set of sim

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::f31ec1519dbca3c93d878e46d93357b7

Enriching how-to guides with actionable phrases and linked data

Autor: Alexandr Chernov, Ágnes Sándor, Matthias Gallé, Nikolaos Lagos

Publikováno v: Web Intelligence. 15:189-203

The World Wide Web contains a large quantity of community created knowledge of instructional nature. Similarly, in a commercial setting, databases of instructions are used by customer-care providers to guide clients in the resolution of issues. Most

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::8b24f2fe6bf59c6735fa35b7c1363cea
https://doi.org/10.3233/web-170364

Joint Semantic and Distributional Word Representations with Multi-Graph Embeddings

Autor: Pierre Daix-Moreux, Matthias Gallé

Publikováno v: TextGraphs@EMNLP

Word embeddings continue to be of great use for NLP researchers and practitioners due to their training speed and easiness of use and distribution. Prior work has shown that the representation of those words can be improved by the use of semantic kno

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::b4f9f9e903f351aad27623512e476da9
https://doi.org/10.18653/v1/d19-5314