Pipeline Analysis for Developing Instruct LLMs in Low-Resource Languages: A Case Study on Basque
Large language models (LLMs) are typically optimized for resource-rich languages like English, exacerbating the gap between high-resource and underrepresented languages. This work presents a detailed analysis of strategies for developing a model capable of following instructions in a low-resource language, specifically Basque, by focusing on three key stages: pre-training, instruction tuning, and alignment with human preferences. Our findings demonstrate that continual pre-training with a high-quality Basque corpus of around 600 million words improves natural language understanding (NLU) of the foundational model by over 12 points. Moreover, instruction tuning and human preference alignment using automatically translated datasets proved highly effective, resulting in a 24-point improvement in instruction-following performance. The resulting models, Llama-eus-8B and Llama-eus-8B-instruct, establish a new state-of-the-art for Basque in the sub-10B parameter category.
Code (0)
등록된 구현이 없습니다.
Tasks
Instruction FollowingNatural Language UnderstandingSimilar Papers 제목 키워드 기반
Translate, then Detect: Leveraging Machine Translation for Cross-Lingual Toxicity Classification
Multilingual toxicity detection remains a significant challenge due to the scarcity of training data and resources for many languages. While prior work has leveraged the translate-test paradigm to support cross-lingual t…
Cross-Lingual TransferMachine TranslationLLMs for Extremely Low-Resource Finno-Ugric Languages
The advancement of large language models (LLMs) has predominantly focused on high-resource languages, leaving low-resource languages, such as those in the Finno-Ugric family, significantly underrepresented. This paper ad…
DiversityToward General Instruction-Following Alignment for Retrieval-Augmented Generation
Following natural instructions is crucial for the effective application of Retrieval-Augmented Generation (RAG) systems. Despite recent advancements in Large Language Models (LLMs), research on assessing and improving in…
Instruction FollowingRAGRetrievalRetrieval-augmented GenerationVikhr: Constructing a State-of-the-art Bilingual Open-Source Instruction-Following Large Language Model for Russian
There has been a surge in developing various Large Language Models (LLMs). However, text generation for languages other than English often faces significant challenges, including poor generation quality and reduced compu…
Instruction FollowingLanguage ModelingLanguage ModellingLarge Language Model+1Benchmarking Hindi LLMs: A New Suite of Datasets and a Comparative Analysis
Evaluating instruction-tuned Large Language Models (LLMs) in Hindi is challenging due to a lack of high-quality benchmarks, as direct translation of English datasets fails to capture crucial linguistic and cultural nuanc…