InstaTrans: An Instruction-Aware Translation Framework for Non-English Instruction Datasets

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kim, Yungi, Park, Chanjun
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912055698653184
author Kim, Yungi
Park, Chanjun
author_facet Kim, Yungi
Park, Chanjun
contents It is challenging to generate high-quality instruction datasets for non-English languages due to tail phenomena, which limit performance on less frequently observed data. To mitigate this issue, we propose translating existing high-quality English instruction datasets as a solution, emphasizing the need for complete and instruction-aware translations to maintain the inherent attributes of these datasets. We claim that fine-tuning LLMs with datasets translated in this way can improve their performance in the target language. To this end, we introduces a new translation framework tailored for instruction datasets, named InstaTrans (INSTruction-Aware TRANSlation). Through extensive experiments, we demonstrate the superiority of InstaTrans over other competitors in terms of completeness and instruction-awareness of translation, highlighting its potential to broaden the accessibility of LLMs across diverse languages at a relatively low cost. Furthermore, we have validated that fine-tuning LLMs with datasets translated by InstaTrans can effectively improve their performance in the target language.
format Preprint
id arxiv_https___arxiv_org_abs_2410_01512
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle InstaTrans: An Instruction-Aware Translation Framework for Non-English Instruction Datasets
Kim, Yungi
Park, Chanjun
Computation and Language
Artificial Intelligence
It is challenging to generate high-quality instruction datasets for non-English languages due to tail phenomena, which limit performance on less frequently observed data. To mitigate this issue, we propose translating existing high-quality English instruction datasets as a solution, emphasizing the need for complete and instruction-aware translations to maintain the inherent attributes of these datasets. We claim that fine-tuning LLMs with datasets translated in this way can improve their performance in the target language. To this end, we introduces a new translation framework tailored for instruction datasets, named InstaTrans (INSTruction-Aware TRANSlation). Through extensive experiments, we demonstrate the superiority of InstaTrans over other competitors in terms of completeness and instruction-awareness of translation, highlighting its potential to broaden the accessibility of LLMs across diverse languages at a relatively low cost. Furthermore, we have validated that fine-tuning LLMs with datasets translated by InstaTrans can effectively improve their performance in the target language.
title InstaTrans: An Instruction-Aware Translation Framework for Non-English Instruction Datasets
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2410.01512