TY - GEN
T1 - KBioXLM: A Knowledge-anchored Biomedical Multilingual Pretrained Language Model
AU - Geng, Lei
AU - Yan, Xu
AU - Cao, Ziqiang
AU - Li, Juntao
AU - Li, Wenjie
AU - Li, Sujian
AU - Zhou, Xinjie
AU - Yang, Yang
AU - Zhang, Jun
N1 - Publisher Copyright:
© 2023 Association for Computational Linguistics.
PY - 2023/12
Y1 - 2023/12
N2 - Most biomedical pretrained language models are monolingual and cannot handle the growing cross-lingual requirements. The scarcity of non-English domain corpora, not to mention parallel data, poses a significant hurdle in training multilingual biomedical models. Since knowledge forms the core of domain-specific corpora and can be translated into various languages accurately, we propose a model called KBioXLM, which transforms the multilingual pretrained model XLM-R into the biomedical domain using a knowledge-anchored approach. We achieve a biomedical multilingual corpus by incorporating three granularity knowledge alignments (entity, fact, and passage levels) into monolingual corpora. Then we design three corresponding training tasks (entity masking, relation masking, and passage relation prediction) and continue training on top of the XLM-R model to enhance its domain cross-lingual ability. To validate the effectiveness of our model, we translate the English benchmarks of multiple tasks into Chinese. Experimental results demonstrate that our model significantly outperforms monolingual and multilingual pretrained models in cross-lingual zero-shot and few-shot scenarios, achieving improvements of up to 10+ points. Our code is publicly available at https://github.com/ngwlh-gl/KBioXLM.
AB - Most biomedical pretrained language models are monolingual and cannot handle the growing cross-lingual requirements. The scarcity of non-English domain corpora, not to mention parallel data, poses a significant hurdle in training multilingual biomedical models. Since knowledge forms the core of domain-specific corpora and can be translated into various languages accurately, we propose a model called KBioXLM, which transforms the multilingual pretrained model XLM-R into the biomedical domain using a knowledge-anchored approach. We achieve a biomedical multilingual corpus by incorporating three granularity knowledge alignments (entity, fact, and passage levels) into monolingual corpora. Then we design three corresponding training tasks (entity masking, relation masking, and passage relation prediction) and continue training on top of the XLM-R model to enhance its domain cross-lingual ability. To validate the effectiveness of our model, we translate the English benchmarks of multiple tasks into Chinese. Experimental results demonstrate that our model significantly outperforms monolingual and multilingual pretrained models in cross-lingual zero-shot and few-shot scenarios, achieving improvements of up to 10+ points. Our code is publicly available at https://github.com/ngwlh-gl/KBioXLM.
UR - http://www.scopus.com/inward/record.url?scp=85183303497&partnerID=8YFLogxK
M3 - Conference article published in proceeding or book
AN - SCOPUS:85183303497
T3 - Findings of the Association for Computational Linguistics: EMNLP 2023
SP - 11239
EP - 11250
BT - Findings of the Association for Computational Linguistics
PB - Association for Computational Linguistics (ACL)
T2 - 2023 Findings of the Association for Computational Linguistics: EMNLP 2023
Y2 - 6 December 2023 through 10 December 2023
ER -