# Sch"utzen: Evaluating LLM Safety in Bulgarian and German Contexts

arXiv: https://arxiv.org/abs/2606.11316 · HTML: https://mlsanity.com/paper/2606.11316/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Safety alignment, refusal and guardrails, one of 12 directions in LLM safety and security.
- Builds on 6 of its 18 references that are in the map.
- First posted on arXiv 9 Jun 2026.

## Placed in

- LLM safety and security / [Safety alignment, refusal and guardrails](https://mlsanity.com/safety/safety-alignment-refusal-and-guardrails/)

## Builds on (most cited first, 6 of 6)

- [EXAMS: A Multi-Subject High School Examinations Dataset for Cross-Lingual and Multilingual Question Answering](https://mlsanity.com/paper/2011.03080/) (2020)
- [Against The Achilles' Heel: A Survey on Red Teaming for Generative Models](https://mlsanity.com/paper/2404.00629/) (2024)
- [Arabic Dataset for LLM Safeguard Evaluation](https://mlsanity.com/paper/2410.17040/) (2024)
- [LL"aMmlein: Transparent, Compact and Competitive German-Only Language Models from Scratch](https://mlsanity.com/paper/2411.11171/) (2024)
- [BgGPT 1.0: Extending English-centric LLMs to other languages](https://mlsanity.com/paper/2412.10893/) (2024)
- [bgGLUE: A Bulgarian General Language Understanding Evaluation Benchmark](https://mlsanity.com/paper/2306.02349/) (2023)

## Abstract (arXiv)

Large language models are increasingly deployed across professional domains, bringing hard-to-predict risks, including the generation of harmful or disrespectful content. Although substantial progress has been made in developing safety evaluation datasets, existing resources remain overwhelmingly English- and Chinese-centric. This limitation is particularly pronounced when evaluating languages that operate within shared sociocultural, legal, and ethical contexts. To address this gap, we introduce Sch"utzen: a German--Bulgarian safety dataset designed to assess model answerability under risk, covering both a low-resource language (Bulgarian) and a high-resource language (German). Experiments with multilingual and language-specific LLMs reveal pronounced cross-language differences in safety behavior, highlighting the necessity of tailored, region-specific evaluation resources to support the responsible deployment of LLMs in Germany and Bulgaria. Datasets and code are available at https://github.com/xnlp-lab/Schutzen. Warning: this paper contains examples that may be offensive, harmful, or biased.