Tracing Multilingual Factual Knowledge Acquisition in Pretraining
MCML Authors
Mingyang Wang
Amir Hossein Kargaran
Felicia Körner
Barbara Plank
Prof. Dr.
Principal Investigator
Hinrich Schütze
Prof. Dr.
Principal Investigator
Abstract
Mingyang Wang
Amir Hossein Kargaran
Felicia Körner
Barbara Plank
Prof. Dr.
Principal Investigator
Hinrich Schütze
Prof. Dr.
Principal Investigator
Abstract
Large Language Models (LLMs) are capable of recalling multilingual factual knowledge present in their pretraining data. However, most studies evaluate only the final model, leaving the development of factual recall and crosslingual consistency throughout pretraining largely unexplored. In this work, we trace how factual recall and crosslingual consistency evolve during pretraining, focusing on OLMo-7B as a case study. We find that both accuracy and consistency improve over time for most languages. We show that this improvement is primarily driven by the fact frequency in the pretraining corpus: more frequent facts are more likely to be recalled correctly, regardless of language. Yet, some low-frequency facts in non-English languages can still be correctly recalled. Our analysis reveals that these instances largely benefit from crosslingual transfer of their English counterparts -- an effect that emerges predominantly in the early stages of pretraining. We pinpoint two distinct pathways through which multilingual factual knowledge acquisition occurs: (1) frequency-driven learning, which is dominant and language-agnostic, and (2) crosslingual transfer, which is limited in scale and typically constrained to relation types involving named entities.
inproceedings
Findings @EMNLP 2025
Findings of the Conference on Empirical Methods in Natural Language Processing. Suzhou, China, Nov 04-09, 2025. To be published. Preprint available.
Authors
Y. Liu • M. Wang • A. H. Kargaran • F. Körner • E. Nie • B. Plank • F. Yvon • H. SchützeLinks
GitHubResearch Area
BibTeXKey: LWK+25a