Improving In-context Learning of Multilingual Generative Language Models with Cross-lingual Alignment

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Chong, Wang, Shaonan, Zhang, Jiajun, Zong, Chengqing
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916283266629632
author Li, Chong
Wang, Shaonan
Zhang, Jiajun
Zong, Chengqing
author_facet Li, Chong
Wang, Shaonan
Zhang, Jiajun
Zong, Chengqing
contents Multilingual generative models obtain remarkable cross-lingual in-context learning capabilities through pre-training on large-scale corpora. However, they still exhibit a performance bias toward high-resource languages and learn isolated distributions of multilingual sentence representations, which may hinder knowledge transfer across languages. To bridge this gap, we propose a simple yet effective cross-lingual alignment framework exploiting pairs of translation sentences. It aligns the internal sentence representations across different languages via multilingual contrastive learning and aligns outputs by following cross-lingual instructions in the target language. Experimental results show that even with less than 0.1 {\textperthousand} of pre-training tokens, our alignment framework significantly boosts the cross-lingual abilities of generative language models and mitigates the performance gap. Further analyses reveal that it results in a better internal multilingual representation distribution of multilingual models.
format Preprint
id arxiv_https___arxiv_org_abs_2311_08089
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Improving In-context Learning of Multilingual Generative Language Models with Cross-lingual Alignment
Li, Chong
Wang, Shaonan
Zhang, Jiajun
Zong, Chengqing
Computation and Language
Multilingual generative models obtain remarkable cross-lingual in-context learning capabilities through pre-training on large-scale corpora. However, they still exhibit a performance bias toward high-resource languages and learn isolated distributions of multilingual sentence representations, which may hinder knowledge transfer across languages. To bridge this gap, we propose a simple yet effective cross-lingual alignment framework exploiting pairs of translation sentences. It aligns the internal sentence representations across different languages via multilingual contrastive learning and aligns outputs by following cross-lingual instructions in the target language. Experimental results show that even with less than 0.1 {\textperthousand} of pre-training tokens, our alignment framework significantly boosts the cross-lingual abilities of generative language models and mitigates the performance gap. Further analyses reveal that it results in a better internal multilingual representation distribution of multilingual models.
title Improving In-context Learning of Multilingual Generative Language Models with Cross-lingual Alignment
topic Computation and Language
url https://arxiv.org/abs/2311.08089