# A Modular Taxonomy for Hate Speech Definitions and Its Impact on Zero-Shot LLM Classification Performance

arXiv: https://arxiv.org/abs/2506.18576 · HTML: https://mlsanity.com/paper/2506.18576/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Placed in Safety alignment, refusal and guardrails, one of 12 directions in LLM safety and security.
- Builds on 24 of its 40 references that are in the map.
- First posted on arXiv 23 Jun 2025.

## Placed in

- LLM safety and security / [Safety alignment, refusal and guardrails](https://mlsanity.com/safety/safety-alignment-refusal-and-guardrails/)

## Builds on (most cited first, 24 of 24)

- [Super-NaturalInstructions: Generalization via Declarative Instructions on 1600+ NLP Tasks](https://mlsanity.com/paper/2204.07705/) (2022)
- [Automated Hate Speech Detection and the Problem of Offensive Language](https://mlsanity.com/paper/1703.04009/) (2017)
- [Instruction Tuning for Large Language Models: A Survey](https://mlsanity.com/paper/2308.10792/) (2023)
- [Large Scale Crowdsourcing and Characterization of Twitter Abusive Behavior](https://mlsanity.com/paper/1802.00393/) (2018)
- [Learning from the Worst: Dynamically Generated Datasets to Improve Online Hate Detection](https://mlsanity.com/paper/2012.15761/) (2020)
- [Latent Hatred: A Benchmark for Understanding Implicit Hate Speech](https://mlsanity.com/paper/2109.05322/) (2021)
- [HateCheck: Functional Tests for Hate Speech Detection Models](https://mlsanity.com/paper/2012.15606/) (2020)
- [Hate Speech Dataset from a White Supremacy Forum](https://mlsanity.com/paper/1809.04444/) (2018)
- [MultiVerS: Improving scientific claim verification with weak supervision and full-document context](https://mlsanity.com/paper/2112.01640/) (2021)
- [Measuring the Reliability of Hate Speech Annotations: The Case of the European Refugee Crisis](https://mlsanity.com/paper/1701.08118/) (2017)
- [Sensitivity, Performance, Robustness: Deconstructing the Effect of Sociodemographic Prompting](https://mlsanity.com/paper/2309.07034/) (2023)
- [Probing LLMs for hate speech detection: strengths and vulnerabilities](https://mlsanity.com/paper/2310.12860/) (2023)
- [Efficient Prompting Methods for Large Language Models: A Survey](https://mlsanity.com/paper/2404.01077/) (2024)
- ["HOT" ChatGPT: The promise of ChatGPT in detecting and discriminating hateful, offensive, and toxic comments on social media](https://mlsanity.com/paper/2304.10619/) (2023)
- [Confronting Abusive Language Online: A Survey from the Ethical and Human Rights Perspective](https://mlsanity.com/paper/2012.12305/) (2020)
- [Mind the instructions: a holistic evaluation of consistency and interactions in prompt-based learning](https://mlsanity.com/paper/2310.13486/) (2023)
- [HARE: Explainable Hate Speech Detection with Step-by-Step Reasoning](https://mlsanity.com/paper/2311.00321/) (2023)
- [People Make Better Edits: Measuring the Efficacy of LLM-Generated Counterfactually Augmented Data for Harmful Language Detection](https://mlsanity.com/paper/2311.01270/) (2023)
- [Hate Speech Criteria: A Modular Approach to Task-Specific Hate Speech Definitions](https://mlsanity.com/paper/2206.15455/) (2022)
- [Recognizing Explicit and Implicit Hate Speech Using a Weakly Supervised Two-path Bootstrapping Approach](https://mlsanity.com/paper/1710.07394/) (2017)
- [Overview of the HASOC Subtrack at FIRE 2021: Hate Speech and Offensive Content Identification in English and Indo-Aryan Languages](https://mlsanity.com/paper/2112.09301/) (2021)
- [Improving Adversarial Data Collection by Supporting Annotators: Lessons from GAHD, a German Hate Speech Dataset](https://mlsanity.com/paper/2403.19559/) (2024)
- [CoSyn: Detecting Implicit Hate Speech in Online Conversations Using a Context Synergized Hyperbolic Network](https://mlsanity.com/paper/2303.03387/) (2023)
- [Weigh Your Own Words: Improving Hate Speech Counter Narrative Generation via Attention Regularization](https://mlsanity.com/paper/2309.02311/) (2023)

## Built on by (most cited first, 7 of 7)

- [Ambiguity Collapse by LLMs: A Taxonomy of Epistemic Risks](https://mlsanity.com/paper/2603.05801/) (2026)
- [xList-Hate: A Checklist-Based Framework for Interpretable and Generalizable Hate Speech Detection](https://mlsanity.com/paper/2602.05874/) (2026)
- [Do We Still Need Humans in the Loop? Human vs. LLM Annotation in Active Learning for TikTok Hate Speech Detection](https://mlsanity.com/paper/2604.13899/) (2026)
- [iPOE: Interpretable Prompt Optimization via Explanations](https://mlsanity.com/paper/2605.18113/) (2026)
- [You Frame It: How Conceptual Representations Shape LLM Detection and Reasoning about Antisemitism](https://mlsanity.com/paper/2607.04945/) (2026)
- ['Ghaib in Translation' aka Unseen Harm: Measuring Cross-Script Safety Inconsistency with 'Missed-in-Urdu' Scores in LLM Hate Speech Detection](https://mlsanity.com/paper/2608.24191/) (2026)
- [To Jev or Not? Evaluating the Accuracy and Efficiency of Structured Decision Models for Hate-Speech Moderation](https://mlsanity.com/paper/2610.03324/) (2026)

## Abstract (arXiv)

Detecting harmful content is a crucial task in the landscape of NLP applications for Social Good, with hate speech being one of its most dangerous forms. But what do we mean by hate speech, how can we define it, and how does prompting different definitions of hate speech affect model performance? The contribution of this work is twofold. At the theoretical level, we address the ambiguity surrounding hate speech by collecting and analyzing existing definitions from the literature. We organize these definitions into a taxonomy of 14 Conceptual Elements-building blocks that capture different aspects of hate speech definitions, such as references to the target of hate (individual or groups) or of the potential consequences of it. At the experimental level, we employ the collection of definitions in a systematic zero-shot evaluation of three LLMs, on three hate speech datasets representing different types of data (synthetic, human-in-the-loop, and real-world). We find that choosing different definitions, i.e., definitions with a different degree of specificity in terms of encoded elements, impacts model performance, but this effect is not consistent across all architectures.