TIBSTC-CoT: A Multi-Domain Instruction Dataset for Chain-of-Thought Reasoning in Language Models

TL;DR

This paper introduces TIBSTC-CoT, a large-scale Tibetan dataset created with LLMs for chain-of-thought reasoning, and develops Tibetan-centric LLMs trained on this dataset, improving reasoning and generation in Tibetan language processing.

Contribution

The paper presents the first large-scale Tibetan reasoning dataset and a series of Tibetan-centric LLMs trained on it, advancing low-resource language AI capabilities.

Findings

Sunshine-thinking LLMs achieve reasoning performance comparable to SOTA multilingual models.

TIBSTC-CoT enables scalable dataset creation for low-resource languages.

The approach promotes inclusive AI for Tibetan language processing.

Abstract

To address the severe data scarcity in Tibetan, a low-resource language spoken by over six million people, we introduce TIBSTC-CoT, the large-scale, multi-domain Tibetan dataset automatically constructed via chain-of-thought prompting with large language models (LLMs). TIBSTC-CoT establishes a scalable and reproducible framework for dataset creation in low-resource settings, covering diverse domains and reasoning patterns essential for language understanding and generation. Building on this dataset, we develop the Sunshine-thinking LLM family, a series of Tibetan-centric LLMs equipped with chain-of-thought capabilities. Trained entirely on TIBSTC-CoT, Sunshine-thinking has demonstrated strong reasoning and generation performance, comparable to state-of-the-art (SOTA) multilingual LLMs. Our work marks a significant step toward inclusive AI by enabling high-quality Tibetan language processing through both resource creation and model innovation. All data are available: https://github.com/Vicentvankor/sun-shine.