RoQLlama: A Lightweight Romanian Adapted Language Model

George-Andrei Dima, Andrei-Marius Avram, Cristian-George Craciun, Dumitru-Clementin Cercel


Abstract
The remarkable achievements obtained by open-source large language models (LLMs) in recent years have predominantly been concentrated on tasks involving the English language. In this paper, we aim to advance the performance of Llama2 models on Romanian tasks. We tackle the problem of reduced computing resources by using QLoRA for training. We release RoQLlama-7b, a quantized LLM, which shows equal or improved results compared to its full-sized counterpart when tested on seven Romanian downstream tasks in the zero-shot setup. Also, it consistently achieves higher average scores across all few-shot prompts. Additionally, we introduce a novel Romanian dataset, namely RoMedQA, which contains single-choice medical questions in Romanian.
Anthology ID:
2024.findings-emnlp.261
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2024
Month:
November
Year:
2024
Address:
Miami, Florida, USA
Editors:
Yaser Al-Onaizan, Mohit Bansal, Yun-Nung Chen
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
4531–4541
Language:
URL:
https://aclanthology.org/2024.findings-emnlp.261
DOI:
Bibkey:
Cite (ACL):
George-Andrei Dima, Andrei-Marius Avram, Cristian-George Craciun, and Dumitru-Clementin Cercel. 2024. RoQLlama: A Lightweight Romanian Adapted Language Model. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 4531–4541, Miami, Florida, USA. Association for Computational Linguistics.
Cite (Informal):
RoQLlama: A Lightweight Romanian Adapted Language Model (Dima et al., Findings 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.findings-emnlp.261.pdf