Pith. sign in

REVIEW 1 cited by

HateBench: Benchmarking Hate Speech Detectors on LLM-Generated Content and Hate Campaigns

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2501.16750 v1 pith:UU5SL5MX submitted 2025-01-28 cs.CR cs.LG

classification cs.CRcs.LG
keywords hatespeechdetectorsllm-generatedattackllmsattackscampaigns
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
abstract

Large Language Models (LLMs) have raised increasing concerns about their misuse in generating hate speech. Among all the efforts to address this issue, hate speech detectors play a crucial role. However, the effectiveness of different detectors against LLM-generated hate speech remains largely unknown. In this paper, we propose HateBench, a framework for benchmarking hate speech detectors on LLM-generated hate speech. We first construct a hate speech dataset of 7,838 samples generated by six widely-used LLMs covering 34 identity groups, with meticulous annotations by three labelers. We then assess the effectiveness of eight representative hate speech detectors on the LLM-generated dataset. Our results show that while detectors are generally effective in identifying LLM-generated hate speech, their performance degrades with newer versions of LLMs. We also reveal the potential of LLM-driven hate campaigns, a new threat that LLMs bring to the field of hate speech detection. By leveraging advanced techniques like adversarial attacks and model stealing attacks, the adversary can intentionally evade the detector and automate hate campaigns online. The most potent adversarial attack achieves an attack success rate of 0.966, and its attack efficiency can be further improved by $13-21\times$ through model stealing attacks with acceptable attack performance. We hope our study can serve as a call to action for the research community and platform moderators to fortify defenses against these emerging threats.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Are Today's LLMs Ready to Explain Well-Being Concepts?

    cs.CL 2025-08 unverdicted novelty 5.0 of 10

    AI judges can score explanations of well-being concepts, and small models fine-tuned with preference data score better than larger models, although judges and explainers are all AIs.

Pith tools