Improved Multilingual Language Model Pretraining for Social Media Text via Translation Pair Prediction
share
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 F1 across target languages) and sentiment classification (12 improvement in F1) on social media text, while also benchmarking on a non-social media task of Universal Dependency POS tagging (6.7 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.
READ FULL TEXT
