UCCIX: Irish-eXcellence Large Language Model

TL;DR

This paper introduces UCCIX, an open-source Irish LLM based on Llama 2-13B, developed with a novel low-resource adaptation framework, outperforming larger models on Irish tasks and providing new benchmarking datasets.

Contribution

The paper presents a novel framework for low-resource language adaptation of LLMs and introduces Irish-specific datasets, advancing Irish language AI capabilities.

Findings

UCCIX outperforms larger models on Irish tasks with up to 12% improvement.

The framework requires only a fraction of typical training data for low-resource languages.

New Irish benchmarking datasets enable rigorous evaluation of Irish LLMs.

Abstract

The development of Large Language Models (LLMs) has predominantly focused on high-resource languages, leaving extremely low-resource languages like Irish with limited representation. This work presents UCCIX, a pioneering effort on the development of an open-source Irish-based LLM. We propose a novel framework for continued pre-training of LLMs specifically adapted for extremely low-resource languages, requiring only a fraction of the textual data typically needed for training LLMs according to scaling laws. Our model, based on Llama 2-13B, outperforms much larger models on Irish language tasks with up to 12% performance improvement, showcasing the effectiveness and efficiency of our approach. We also contribute comprehensive Irish benchmarking datasets, including IrishQA, a question-answering dataset, and Irish version of MT-bench. These datasets enable rigorous evaluation and facilitate future research in Irish LLM systems. Our work aims to preserve and promote the Irish language, knowledge, and culture of Ireland in the digital era while providing a framework for adapting LLMs to other indigenous languages.