When Does Hate Transfer? A Large-Scale Study of Cross-Lingual Hate Speech Detection

Key points

• This work investigates a core practical question: Can training data from one language support hate-speech detection in another language with comparable performance?1
• To answer this question, we conduct a large-scale study across 14 languages. \change{We observe consistent patterns in which certain language pairs transfer more effectively than others.1
• We present this cultural interpretation as a hypothesis from qualitative inspection rather than a measured cause.}{} Our analysis includes low-resource languages, offering practical insights into when cross-lingual transfer can support them and where its limitations emerge.1
• These findings inform the design of more reliable multilingual hate-detection systems.1

Sources (1)

• [1]When Does Hate Transfer? A Large-Scale Study of Cross-Lingual Hate Speech Detection
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 19, 08:58 PM

  Effective detection of hate content is essential for maintaining healthy online communities, yet performance can vary substantially across languages.

  This work investigates a core practical question: Can training data from one language support hate-speech detection in another language with comparable performance?

Extractive summary: sentences quoted from the sources.