Cross-lingual QA: A Key to Unlocking In-context Cross-lingual Performance

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kim, Sunkyoung, Ki, Dayeon, Kim, Yireun, Lee, Jinsik
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911956139507712
author Kim, Sunkyoung
Ki, Dayeon
Kim, Yireun
Lee, Jinsik
author_facet Kim, Sunkyoung
Ki, Dayeon
Kim, Yireun
Lee, Jinsik
contents Multilingual large language models (MLLMs) have demonstrated significant cross-lingual capabilities through in-context learning. Existing approaches typically construct monolingual in-context examples, either in the source or target language. However, translating entire in-context examples into the target language might compromise contextual integrity and be costly in the case of long-context passages. To address this, we introduce Cross-lingual QA, a cross-lingual prompting method that translates only the question and answer parts, thus reducing translation costs. Experiments on four typologically diverse multilingual benchmarks show that Cross-lingual QA prompting effectively stimulates models to elicit their cross-lingual knowledge, outperforming prior monolingual prompting approaches. Furthermore, we show that prompting open-source MLLMs with cross-lingual in-context examples enhances performance as the model scale increases.
format Preprint
id arxiv_https___arxiv_org_abs_2305_15233
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Cross-lingual QA: A Key to Unlocking In-context Cross-lingual Performance
Kim, Sunkyoung
Ki, Dayeon
Kim, Yireun
Lee, Jinsik
Computation and Language
Artificial Intelligence
Multilingual large language models (MLLMs) have demonstrated significant cross-lingual capabilities through in-context learning. Existing approaches typically construct monolingual in-context examples, either in the source or target language. However, translating entire in-context examples into the target language might compromise contextual integrity and be costly in the case of long-context passages. To address this, we introduce Cross-lingual QA, a cross-lingual prompting method that translates only the question and answer parts, thus reducing translation costs. Experiments on four typologically diverse multilingual benchmarks show that Cross-lingual QA prompting effectively stimulates models to elicit their cross-lingual knowledge, outperforming prior monolingual prompting approaches. Furthermore, we show that prompting open-source MLLMs with cross-lingual in-context examples enhances performance as the model scale increases.
title Cross-lingual QA: A Key to Unlocking In-context Cross-lingual Performance
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2305.15233