aiqbench analysis
This research introduces SarcasTraps, a specialized dataset designed to evaluate how sarcasm impacts hate speech detection models in Turkish. The study reveals that while fine-tuned encoders fail to identify sarcastic hate speech, large language models maintain recall at the cost of significant precision drops on these nuanced examples. Ultimately, the paper provides critical quantitative evidence that current NLP systems struggle with indirect harm, highlighting a persistent gap between explicit and implicit toxicity detection.
SarcasTraps: Effect of Sarcasm on Turkish Hate Speech Detection
arXiv:2610.16957v1 Announce Type: cross Abstract: Standard hate speech detection datasets often emphasize explicit harmful language, but directed harm can also be expressed indirectly, without clearly hateful wording. Sarcasm is one such form of indirect expression, as it can convey harmful intent through mock praise, exaggeration, humor, or apparent politeness. In this paper, we introduce SarcasTraps, a targeted external test set for analyzing the effect of sarcasm on Turkish hate speech detection. SarcasTraps contains 479 entries annotated along two dimensions: hate speech and sarcasm. We train classifiers using BERTurk and XLM-RoBERTa on two existing Turkish hate speech datasets and evaluate them on SarcasTraps, using it strictly as an external test set. We also evaluate GPT-4.1 and Llama-3.3-70B in a zero-shot setting on SarcasTraps. Results show that sarcasm substantially affects hate speech detection. Fine-tuned encoder models lose recall on sarcastic hate, while zero-shot LLMs maintain higher recall but suffer substantial precision drops on sarcastic examples. These findings show that both fine-tuned hate speech detectors and zero-shot LLMs struggle when sarcasm is involved, and SarcasTraps provides quantitative evidence of this challenge through a targeted Turkish test set. Due to the presence of sensitive and hateful language, SarcasTraps will be shared upon request for academic research purposes. Warning: This paper has examples from the dataset which contains offensive and hateful content towards groups of people.
Abstract:Standard hate speech detection datasets often emphasize explicit harmful language, but directed harm can also be expressed indirectly, without clearly hateful wording. Sarcasm is one such form of indirect expression, as it can convey harmful intent through mock praise, exaggeration, humor, or apparent politeness. In this paper, we introduce SarcasTraps, a targeted external test set for analyzing the effect of sarcasm on Turkish hate speech detection. SarcasTraps contains 479 entries annotated along two dimensions: hate speech and sarcasm. We train classifiers using BERTurk and XLM-RoBERTa on two existing Turkish hate speech datasets and evaluate them on SarcasTraps, using it strictly as an external test set. We also evaluate GPT-4.1 and Llama-3.3-70B in a zero-shot setting on SarcasTraps. Results show that sarcasm substantially affects hate speech detection. Fine-tuned encoder models lose recall on sarcastic hate, while zero-shot LLMs maintain higher recall but suffer substantial precision drops on sarcastic examples. These findings show that both fine-tuned hate speech detectors and zero-shot LLMs struggle when sarcasm is involved, and SarcasTraps provides quantitative evidence of this challenge through a targeted Turkish test set. Due to the presence of sensitive and hateful language, SarcasTraps will be shared upon request for academic research purposes. Warning: This paper has examples from the dataset which contains offensive and hateful content towards groups of people.
Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
arXiv:2610.16957 [cs.CL]
(or arXiv:2610.16957v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2610.16957 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 317 words
Related coverage
Dealing with Annotator Disagreement in Hate Speech Classification
arXiv · Jun 17, 2026
Dealing with Annotator Disagreement in Hate Speech Classification
arXiv · Jun 9, 2026
SAFE-MEME: Structured Reasoning Framework for Robust Hate Speech Detection in Memes
arXiv · Jul 28, 2026
An Explainable DistilBERT-BiLSTM-Attention Framework for Binary and Multi-Class Hate Speech Detection
arXiv · Sep 25, 2026