Qorgau: Evaluating LLM Safety in Kazakh-Russian Bilingual Contexts

TL;DR

This paper introduces Qorgau, a dataset for evaluating the safety of large language models in Kazakh-Russian bilingual contexts, highlighting the importance of region-specific safety assessments.

Contribution

It presents a novel bilingual safety evaluation dataset for Kazakh and Russian, addressing the gap in safety assessment tools for low-resource and bilingual settings.

Findings

Multilingual and language-specific LLMs show different safety performance.

Region-specific datasets are crucial for responsible LLM deployment.

Bilingual safety evaluation reveals unique risks not seen in monolingual contexts.

Abstract

Large language models (LLMs) are known to have the potential to generate harmful content, posing risks to users. While significant progress has been made in developing taxonomies for LLM risks and safety evaluation prompts, most studies have focused on monolingual contexts, primarily in English. However, language- and region-specific risks in bilingual contexts are often overlooked, and core findings can diverge from those in monolingual settings. In this paper, we introduce Qorgau, a novel dataset specifically designed for safety evaluation in Kazakh and Russian, reflecting the unique bilingual context in Kazakhstan, where both Kazakh (a low-resource language) and Russian (a high-resource language) are spoken. Experiments with both multilingual and language-specific LLMs reveal notable differences in safety performance, emphasizing the need for tailored, region-specific datasets to ensure the responsible and safe deployment of LLMs in countries like Kazakhstan. Warning: this paper contains example data that may be offensive, harmful, or biased.