Saved in:
Bibliographic Details
Main Authors: Zhao, Xin, Yoshinaga, Naoki, Oba, Daisuke
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2403.05189
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909132449120256
author Zhao, Xin
Yoshinaga, Naoki
Oba, Daisuke
author_facet Zhao, Xin
Yoshinaga, Naoki
Oba, Daisuke
contents Acquiring factual knowledge for language models (LMs) in low-resource languages poses a serious challenge, thus resorting to cross-lingual transfer in multilingual LMs (ML-LMs). In this study, we ask how ML-LMs acquire and represent factual knowledge. Using the multilingual factual knowledge probing dataset, mLAMA, we first conducted a neuron investigation of ML-LMs (specifically, multilingual BERT). We then traced the roots of facts back to the knowledge source (Wikipedia) to identify the ways in which ML-LMs acquire specific facts. We finally identified three patterns of acquiring and representing facts in ML-LMs: language-independent, cross-lingual shared and transferred, and devised methods for differentiating them. Our findings highlight the challenge of maintaining consistent factual knowledge across languages, underscoring the need for better fact representation learning in ML-LMs.
format Preprint
id arxiv_https___arxiv_org_abs_2403_05189
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Tracing the Roots of Facts in Multilingual Language Models: Independent, Shared, and Transferred Knowledge
Zhao, Xin
Yoshinaga, Naoki
Oba, Daisuke
Computation and Language
Artificial Intelligence
Acquiring factual knowledge for language models (LMs) in low-resource languages poses a serious challenge, thus resorting to cross-lingual transfer in multilingual LMs (ML-LMs). In this study, we ask how ML-LMs acquire and represent factual knowledge. Using the multilingual factual knowledge probing dataset, mLAMA, we first conducted a neuron investigation of ML-LMs (specifically, multilingual BERT). We then traced the roots of facts back to the knowledge source (Wikipedia) to identify the ways in which ML-LMs acquire specific facts. We finally identified three patterns of acquiring and representing facts in ML-LMs: language-independent, cross-lingual shared and transferred, and devised methods for differentiating them. Our findings highlight the challenge of maintaining consistent factual knowledge across languages, underscoring the need for better fact representation learning in ML-LMs.
title Tracing the Roots of Facts in Multilingual Language Models: Independent, Shared, and Transferred Knowledge
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2403.05189