Harmful language detection is commonly evaluated on broadly defined hate speech and toxicity benchmarks, yet it remains unclear whether models that successfully detect harmful language in general can transfer this capability to more specific forms of hate. We investigate this problem for misogyny detection in Polish, an area for which dedicated resources remain scarce. We introduce a new dataset of Polish social media comments annotated for misogyny, gender-directed hateful speech. Using this resource, we first assess the reliability of LLM-assisted annotation and then evaluate general-to-specific transfer by comparing models trained on general harmful language data, safety-specialized models, and models fine-tuned directly for misogyny detection. Our results show that strong performance on general harmful language detection does not transfer uniformly to misogyny; safety-specialized models demonstrate substantially stronger generalization ability. The study contributes a new Polish resource for gender-directed harmful language and demonstrates that the scope and granularity of prior safety training are important factors in transferring harmful language detection from general to specific forms of hate.
A Dataset for Misogyny Detection in Polish Social Media: Transferability of Hateful Speech Detection from General to Specific Types of Hate
Harmful language detection is commonly evaluated on broadly defined hate speech and toxicity benchmarks, yet it remains unclear whether models that successfully detect harmful language in general can transfer this capability to more specific forms of hate.
- Preview

- Year
- 2026
- Hosting
- Full text hostedCC-BY-4.0
Cite
Notes
Only stored in your browser.
Attribution
- Abstract & full text
- arxiv.org/abs/2611.20044CC-BY-4.0
- TL;DR
- Semantic Scholar