Bridging the Bosphorus: Advancing Turkish Large Language Models through Strategies for Low-Resource Language Adaptation and Benchmarking

TL;DR

This paper investigates strategies for developing effective Turkish Large Language Models by adapting existing models and training new ones, addressing low-resource challenges, and establishing benchmarks to evaluate reasoning and knowledge skills.

Contribution

It introduces a comprehensive approach combining adaptation, new model training, and benchmarking for Turkish LLMs, with insights on data, model scaling, and knowledge transfer.

Findings

Adapted English LLMs improve Turkish understanding.

Developed a new Turkish instruction-tuning dataset.

Created a Turkish LLM leaderboard with reasoning benchmarks.

Abstract

Large Language Models (LLMs) are becoming crucial across various fields, emphasizing the urgency for high-quality models in underrepresented languages. This study explores the unique challenges faced by low-resource languages, such as data scarcity, model selection, evaluation, and computational limitations, with a special focus on Turkish. We conduct an in-depth analysis to evaluate the impact of training strategies, model choices, and data availability on the performance of LLMs designed for underrepresented languages. Our approach includes two methodologies: (i) adapting existing LLMs originally pretrained in English to understand Turkish, and (ii) developing a model from the ground up using Turkish pretraining data, both supplemented with supervised fine-tuning on a novel Turkish instruction-tuning dataset aimed at enhancing reasoning capabilities. The relative performance of these methods is evaluated through the creation of a new leaderboard for Turkish LLMs, featuring benchmarks that assess different reasoning and knowledge skills. Furthermore, we conducted experiments on data and model scaling, both during pretraining and fine-tuning, simultaneously emphasizing the capacity for knowledge transfer across languages and addressing the challenges of catastrophic forgetting encountered during fine-tuning on a different language. Our goal is to offer a detailed guide for advancing the LLM framework in low-resource linguistic contexts, thereby making natural language processing (NLP) benefits more globally accessible.