Prompt-Tuning Can Be Much Better Than Fine-Tuning on Cross-lingual Understanding With Multilingual Language Models

TL;DR

This paper demonstrates that prompt tuning significantly outperforms traditional fine-tuning for cross-lingual transfer in multilingual models, using minimal parameter updates across various NLU tasks.

Contribution

It introduces prompt tuning as a superior alternative to fine-tuning for cross-lingual transfer, with extensive evaluation across multiple tasks and languages.

Findings

Prompt tuning outperforms fine-tuning in cross-lingual transfer.

Only 0.1% to 0.3% parameters are tuned in prompt tuning.

Prompt tuning achieves better aligned decision boundaries.

Abstract

Pre-trained multilingual language models show significant performance gains for zero-shot cross-lingual model transfer on a wide range of natural language understanding (NLU) tasks. Previously, for zero-shot cross-lingual evaluation, pre-trained models are only fine-tuned on English data and tested on a variety of target languages. In this paper, we do cross-lingual evaluation on various NLU tasks (sentence classification, sequence labeling, question answering) using prompt-tuning and compare it with fine-tuning. The results show that prompt tuning achieves much better cross-lingual transfer than fine-tuning across datasets, with only 0.1% to 0.3% tuned parameters. Additionally, we demonstrate through the analysis that prompt tuning can have better cross-lingual transferability of representations on downstream tasks with better aligned decision boundaries.