Hatevolution: What Static Benchmarks Don’t Tell Us
Chiara Di Bonaventura, Barbara McGillivray, Yulan He, Albert Meroño-Peñuela
Abstract
Language changes over time, including in the hate speech domain, which evolves quickly following social dynamics and cultural shifts. While NLP research has investigated the impact of language evolution on model training and has proposed several solutions for it, its impact on model benchmarking remains under-explored. Yet, hate speech benchmarks play a crucial role to ensure model safety. In this paper, we empirically evaluate the robustness of 20 language models across two evolving hate speech experiments, and we show the temporal misalignment between static and time-sensitive evaluations. Our findings call for time-sensitive linguistic benchmarks in order to correctly and reliably evaluate language models in the hate speech domain.
BibTeX
@inproceedings{di-bonaventura-etal-2025-hatevolution,
title = "Hatevolution: What Static Benchmarks Don{'}t Tell Us",
author = "Di Bonaventura, Chiara and
McGillivray, Barbara and
He, Yulan and
Mero{\~n}o-Pe{\~n}uela, Albert",
editor = "Che, Wanxiang and
Nabende, Joyce and
Shutova, Ekaterina and
Pilehvar, Mohammad Taher",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2025",
month = jul,
year = "2025",
address = "Vienna, Austria",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.findings-acl.910/",
doi = "10.18653/v1/2025.findings-acl.910",
pages = "17695--17707",
ISBN = "979-8-89176-256-5"
}