Improved Multilingual Language Model Pretraining for Social Media Text via Translation Pair Prediction

Shubhanshu Mishra, Aria Haghighi


Abstract
We evaluate a simple approach to improving zero-shot multilingual transfer of mBERT on social media corpus by adding a pretraining task called translation pair prediction (TPP), which predicts whether a pair of cross-lingual texts are a valid translation. Our approach assumes access to translations (exact or approximate) between source-target language pairs, where we fine-tune a model on source language task data and evaluate the model in the target language. In particular, we focus on language pairs where transfer learning is difficult for mBERT: those where source and target languages are different in script, vocabulary, and linguistic typology. We show improvements from TPP pretraining over mBERT alone in zero-shot transfer from English to Hindi, Arabic, and Japanese on two social media tasks: NER (a 37% average relative improvement in F1 across target languages) and sentiment classification (12% relative improvement in F1) on social media text, while also benchmarking on a non-social media task of Universal Dependency POS tagging (6.7% relative improvement in accuracy). Our results are promising given the lack of social media bitext corpus. Our code can be found at: https://github.com/twitter-research/multilingual-alignment-tpp.
Anthology ID:
2021.wnut-1.42
Volume:
Proceedings of the Seventh Workshop on Noisy User-generated Text (W-NUT 2021)
Month:
November
Year:
2021
Address:
Online
Editors:
Wei Xu, Alan Ritter, Tim Baldwin, Afshin Rahimi
Venue:
WNUT
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
381–388
Language:
URL:
https://aclanthology.org/2021.wnut-1.42
DOI:
10.18653/v1/2021.wnut-1.42
Bibkey:
Cite (ACL):
Shubhanshu Mishra and Aria Haghighi. 2021. Improved Multilingual Language Model Pretraining for Social Media Text via Translation Pair Prediction. In Proceedings of the Seventh Workshop on Noisy User-generated Text (W-NUT 2021), pages 381–388, Online. Association for Computational Linguistics.
Cite (Informal):
Improved Multilingual Language Model Pretraining for Social Media Text via Translation Pair Prediction (Mishra & Haghighi, WNUT 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.wnut-1.42.pdf
Video:
 https://aclanthology.org/2021.wnut-1.42.mp4
Data
WikiMatrix