Bielik Guard: Efficient Polish Language Safety Classifiers for LLM Content Moderation

TL;DR

Bielik Guard introduces two compact Polish language safety classifiers for LLM content moderation, achieving high accuracy and efficiency, and is publicly available for safer AI deployment.

Contribution

The paper presents a new family of efficient Polish safety classifiers based on fine-tuned models, with superior performance and low false positive rates compared to existing solutions.

Findings

0.5B model achieves F1 scores of 0.791 (micro) and 0.785 (macro).

0.1B model demonstrates exceptional efficiency and low false positive rate.

Models outperform HerBERT-PL-Guard in precision and false positive rate.

Abstract

As Large Language Models (LLMs) become increasingly deployed in Polish language applications, the need for efficient and accurate content safety classifiers has become paramount. We present Bielik Guard, a family of compact Polish language safety classifiers comprising two model variants: a 0.1B parameter model based on MMLW-RoBERTa-base and a 0.5B parameter model based on PKOBP/polish-roberta-8k. Fine-tuned on a community-annotated dataset of 6,885 Polish texts, these models classify content across five safety categories: Hate/Aggression, Vulgarities, Sexual Content, Crime, and Self-Harm. Our evaluation demonstrates that both models achieve strong performance on multiple benchmarks. The 0.5B variant offers the best overall discrimination capability with F1 scores of 0.791 (micro) and 0.785 (macro) on the test set, while the 0.1B variant demonstrates exceptional efficiency. Notably, Bielik Guard 0.1B v1.1 achieves superior precision (77.65%) and very low false positive rate (0.63%) on real user prompts, outperforming HerBERT-PL-Guard (31.55% precision, 4.70% FPR) despite identical model size. The models are publicly available and designed to provide appropriate responses rather than simple content blocking, particularly for sensitive categories like self-harm.