About
I am a second-year PhD student at the Sheffield NLP Group, researching cross-lingual transfer and efficient language modeling. With over seven years of academic and industrial NLP experience, I earned an MSc. in Computer Science with Speech and Language Processing (Distinction) from the University of Sheffield in 2020. From 2021 to 2023, I worked as a researcher in the R&D Group at Hitachi, Ltd. in Japan, gaining practical experience in information extraction and efficient language model development. I have published at top-tier conferences, including EMNLP, ACL, and EACL. I am also actively involved in the NLP community as a member of the ACL Rolling Review (ARR) support team.
My curriculum vitae (CV) is available here (Updated on April 26, 2025).
Research Interests
Cross-lingual Transfer, Language Modelling, Natural Language Understanding, Natural Language Processing, Machine Learning
Recent Work
-
Towards efficient and fair natural language processing across languages
The aim of this research is to extend the applicability of large language models (LLMs) to languages beyond English, making AI technologies more accessible to a global audience. My work focuses on efficient cross-lingual adaptation and vocabulary expansion techniques that improve LLM performance across diverse languages.
- Atsuki Yamaguchi, Terufumi Morishita, Aline Villavicencio and Nikolaos Aletras, “Vocabulary Expansion of Chat Models with Unlabeled Target Language Data,” arXiv preprint: 2412.11704, December 2024
- Atsuki Yamaguchi, Aline Villavicencio and Nikolaos Aletras, “How Can We Effectively Expand the Vocabulary of LLMs with 0.01GB of Target Language Text?,” arXiv preprint: 2406.11477, June 2024. (Last updated in September 2024)
- Atsuki Yamaguchi, Aline Villavicencio and Nikolaos Aletras, “An Empirical Study on Cross-lingual Vocabulary Adaptation for Efficient Language Model Inference,” Findings of the Association for Computational Linguistics: EMNLP 2024, November 2024.
-
Explore simple pretraining alternatives for Transformer-based language representaion models
The aim of this research is to investigate simple yet effective pretraining objectives.
- Atsuki Yamaguchi, Hiroaki Ozaki, Terufumi Morishita, Gaku Morio and Yasuhiro Sogawa, “How does the task complexity of masked pretraining objectives affect downstream performance?,” Findings of the Association for Computational Linguistics: ACL 2023, July 2023. (Short paper)
- Atsuki Yamaguchi, George Chrysostomou, Katerina Margatina and Nikolaos Aletras, “Frustratingly Simple Pretraining Alternatives to Masked Language Modeling,” The 2021 Conference on Empirical Methods in Natural Language Processing (EMNLP 2021), Online, November 2021. (Short paper)
Links
Contact
You can send me messages via this contact form.