French Contextualized Word-Embeddings with a sip of CaBeRnet: a New French Balanced Reference Corpus - Archive ouverte HAL Access content directly
Conference Papers Year : 2020

French Contextualized Word-Embeddings with a sip of CaBeRnet: a New French Balanced Reference Corpus

Abstract

This paper describes and compares the impact of different types and size of training corpora on language models like ELMO. By asking the fundamental question of quality versus quantity we evaluate four French corpora for training on parsing scores, POS-tagging and named-entities recognition downstream tasks. The paper studies the relevance of a new corpus, CaBeRnet, featuring a representative range of language usage, including a balanced variety of genres (oral transcriptions, newspapers, popular magazines, technical reports, fiction, academic texts), in oral and written styles. We hypothesize that a linguistically representative and balanced corpora will allow the language model to be more efficient and representative of a given language and therefore yield better evaluation scores on different evaluation sets and tasks.
Fichier principal
Vignette du fichier
LREC_Fabre_Ortiz.pdf (211.09 Ko) Télécharger le fichier
Origin Files produced by the author(s)
Loading...

Dates and versions

hal-02678358 , version 1 (31-05-2020)

Identifiers

  • HAL Id : hal-02678358 , version 1

Cite

Murielle Fabre, Pedro Javier Ortiz Suárez, Benoît Sagot, Éric Villemonte de La Clergerie. French Contextualized Word-Embeddings with a sip of CaBeRnet: a New French Balanced Reference Corpus. CMLC-8 - 8th Workshop on the Challenges in the Management of Large Corpora, May 2020, Marseille, France. ⟨hal-02678358⟩
230 View
545 Download

Share

Gmail Mastodon Facebook X LinkedIn More