When Being Unseen from mBERT is just the Beginning: Handling New Languages With Multilingual Language Models - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2021

When Being Unseen from mBERT is just the Beginning: Handling New Languages With Multilingual Language Models

Abstract

Transfer learning based on pretraining language models on a large amount of raw data has become a new norm to reach state-of-theart performance in NLP. Still, it remains unclear how this approach should be applied for unseen languages that are not covered by any available large-scale multilingual language model and for which only a small amount of raw data is generally available. In this work, by comparing multilingual and monolingual models, we show that such models behave in multiple ways on unseen languages. Some languages greatly benefit from transfer learning and behave similarly to closely related high resource languages whereas others apparently do not. Focusing on the latter, we show that this failure to transfer is largely related to the impact of the script used to write such languages. We show that transliterating those languages significantly improves the potential of large-scale multilingual language models on downstream tasks. This result provides a promising direction towards making these massively multilingual models useful for a new set of unseen languages.
Fichier principal
Vignette du fichier
NAACL21_Muller_et_al.pdf (281.12 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03251105 , version 1 (03-02-2022)

Identifiers

  • HAL Id : hal-03251105 , version 1

Cite

Benjamin Muller, Antonios Anastasopoulos, Benoît Sagot, Djamé Seddah. When Being Unseen from mBERT is just the Beginning: Handling New Languages With Multilingual Language Models. NAACL-HLT 2021 - 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Jun 2021, Mexico City, Mexico. ⟨hal-03251105⟩
56 View
143 Download

Share

Gmail Facebook X LinkedIn More