About me

I am a fourth-year PhD student at the Natural Language Processing group at the University of Tartu supervised by Prof. Mark Fishel.

My main research focus is on cross-lingual transfer learning to bring NLP advancements to low-resource languages. During my PhD studies, I have explored methods to leverage translated data for improving natural language understanding in low-resource settings. I have also addressed the limitations of this approach. Additionally, I am interested in evaluating LLMs and improving reasoning abilities of language models in low-resource settings. I helped to create Llammas 🐑.

Previously, I earned my Bachelor’s and Master’s degrees in Computer Science from the University of Tartu. During my studies, I completed an internship at STACC, where I worked on detecting adverse drug reactions from patient epicrises, which led to the writing of my Master’s thesis. I have also worked as a data scientist and as a software developer.

News

23/01/2025

LLMs for Extremely Low-Resource Finno-Ugric Languages was accepted at NAACL 2025 Findings.

Selected publications

Taido Purason^†, Hele-Andra Kuulmets^†, Mark Fišel. 2024. LLMs for Extremely Low-Resource Finno-Ugric Languages. Pre-print. [arXiv]
Hele-Andra Kuulmets^†, Taido Purason^†, Agnes Luhtaru, Mark Fišel. 2024. Teaching Llama a New Language Through Cross-Lingual Knowledge Transfer. In Findings of NAACL 2024. [arXiv] [anthology]
Hele-Andra Kuulmets and Mark Fišel. 2023. Translated Benchmarks Can Be Misleading: the Case of Estonian Question Answering. In Proceedings of NoDaLiDa 2023. [anthology]
Hele-Andra Kuulmets, Andre Tättar, Mark Fišel. 2022. Estonian Language Understanding: a Case Study on the COPA Task. In Baltic Journal of Modern Computing. [pdf]

_{^† Equal contribution}