RedWhale: An Adapted Korean LLM Through Efficient Continual Pretraining

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Vo, Anh-Dung, Jung, Minseong, Lee, Wonbeen, Choi, Daewoo
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914918635143168
author Vo, Anh-Dung
Jung, Minseong
Lee, Wonbeen
Choi, Daewoo
author_facet Vo, Anh-Dung
Jung, Minseong
Lee, Wonbeen
Choi, Daewoo
contents The field of Natural Language Processing (NLP) has seen significant advancements with the development of Large Language Models (LLMs). However, much of this research remains focused on English, often overlooking low-resource languages like Korean. This oversight presents challenges due to the unique non-alphabetic token structure of Korean and the substantial memory and computational demands required for LLM training, which frequently lead to memory constraints and out-of-memory errors. To address these issues, we present RedWhale, a model specifically tailored for Korean language processing. RedWhale is developed using an efficient continual pretraining approach that includes a comprehensive Korean corpus preprocessing pipeline, a specialized tokenizer, an optimized model initialization technique, and a multistage pretraining strategy. These innovations collectively reduce training time and computational costs while maintaining high levels of accuracy and comprehension. By leveraging cross-lingual transfer learning, which exploits shared linguistic similarities across languages, RedWhale builds on English models to enhance Korean language processing. Experimental results demonstrate that RedWhale outperforms other leading models on Korean NLP benchmarks, including the Korean Balanced Evaluation of Significant Tasks (KoBEST), showing superior understanding and generation of Korean text. Furthermore, RedWhale showed no signs of convergence even after pretraining on 9.7 billion tokens, indicating the potential for further improvements with additional training. This work represents a significant advancement in bridging the linguistic divide, particularly in enhancing NLP capabilities for the Korean language.
format Preprint
id arxiv_https___arxiv_org_abs_2408_11294
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle RedWhale: An Adapted Korean LLM Through Efficient Continual Pretraining
Vo, Anh-Dung
Jung, Minseong
Lee, Wonbeen
Choi, Daewoo
Computation and Language
The field of Natural Language Processing (NLP) has seen significant advancements with the development of Large Language Models (LLMs). However, much of this research remains focused on English, often overlooking low-resource languages like Korean. This oversight presents challenges due to the unique non-alphabetic token structure of Korean and the substantial memory and computational demands required for LLM training, which frequently lead to memory constraints and out-of-memory errors. To address these issues, we present RedWhale, a model specifically tailored for Korean language processing. RedWhale is developed using an efficient continual pretraining approach that includes a comprehensive Korean corpus preprocessing pipeline, a specialized tokenizer, an optimized model initialization technique, and a multistage pretraining strategy. These innovations collectively reduce training time and computational costs while maintaining high levels of accuracy and comprehension. By leveraging cross-lingual transfer learning, which exploits shared linguistic similarities across languages, RedWhale builds on English models to enhance Korean language processing. Experimental results demonstrate that RedWhale outperforms other leading models on Korean NLP benchmarks, including the Korean Balanced Evaluation of Significant Tasks (KoBEST), showing superior understanding and generation of Korean text. Furthermore, RedWhale showed no signs of convergence even after pretraining on 9.7 billion tokens, indicating the potential for further improvements with additional training. This work represents a significant advancement in bridging the linguistic divide, particularly in enhancing NLP capabilities for the Korean language.
title RedWhale: An Adapted Korean LLM Through Efficient Continual Pretraining
topic Computation and Language
url https://arxiv.org/abs/2408.11294