Online platforms struggle to curb hate speech without over-censoring legitimate discourse. Early bidirectional transformer encoders made big strides, but the arrival of ultra-large autoregressive LLMs promises deeper context-awareness. Whether this extra scale actually improves practical hate-speech detection on real-world text remains unverified. Our study puts this question to the test by benchmarking both model families, classic encoders and next-generation LLMs, on curated corpora of online interactions for hate-speech detection (Hate or No Hate).
@article{arxiv.2507.10468,
title = {From BERT to Qwen: Hate Detection across architectures},
author = {Ariadna Mon and Saúl Fenollosa and Jon Lecumberri},
journal= {arXiv preprint arXiv:2507.10468},
year = {2025}
}