Boosting Cross-Lingual Transfer via Self-Learning with Uncertainty Estimation
This work addresses the challenge of enhancing zero-shot performance in cross-lingual tasks for NLP practitioners, representing an incremental improvement over existing methods.
The paper tackles the problem of improving cross-lingual transfer in multilingual models by proposing a self-learning framework that uses unlabeled target language data with uncertainty estimation to select high-quality silver labels, resulting in significant performance gains of 10 F1 on average for NER and 2.5 accuracy for NLI across 40 languages.
Recent multilingual pre-trained language models have achieved remarkable zero-shot performance, where the model is only finetuned on one source language and directly evaluated on target languages. In this work, we propose a self-learning framework that further utilizes unlabeled data of target languages, combined with uncertainty estimation in the process to select high-quality silver labels. Three different uncertainties are adapted and analyzed specifically for the cross lingual transfer: Language Heteroscedastic/Homoscedastic Uncertainty (LEU/LOU), Evidential Uncertainty (EVI). We evaluate our framework with uncertainties on two cross-lingual tasks including Named Entity Recognition (NER) and Natural Language Inference (NLI) covering 40 languages in total, which outperforms the baselines significantly by 10 F1 on average for NER and 2.5 accuracy score for NLI.