# Improving Adversarial Data Collection by Supporting Annotators: Lessons from GAHD, a German Hate Speech Dataset

arXiv: https://arxiv.org/abs/2403.19559 · HTML: https://mlsanity.com/paper/2403.19559/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- Builds on 31 of its 56 references that are in the map.
- First posted on arXiv 28 Mar 2024.

## Builds on (most cited first, 30 of 31)

- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction](https://mlsanity.com/paper/1802.03426/) (2018)
- [DROP: A Reading Comprehension Benchmark Requiring Discrete Reasoning Over Paragraphs](https://mlsanity.com/paper/1903.00161/) (2019)
- [Adversarial NLI: A New Benchmark for Natural Language Understanding](https://mlsanity.com/paper/1910.14599/) (2019)
- [Making Monolingual Sentence Embeddings Multilingual using Knowledge Distillation](https://mlsanity.com/paper/2004.09813/) (2020)
- [Dynabench: Rethinking Benchmarking in NLP](https://mlsanity.com/paper/2104.14337/) (2021)
- [Learning the Difference that Makes a Difference with Counterfactually-Augmented Data](https://mlsanity.com/paper/1909.12434/) (2019)
- [SWAG: A Large-Scale Adversarial Dataset for Grounded Commonsense Inference](https://mlsanity.com/paper/1808.05326/) (2018)
- [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)
- [HateCheck: Functional Tests for Hate Speech Detection Models](https://mlsanity.com/paper/2012.15606/) (2020)
- [Build it Break it Fix it for Dialogue Safety: Robustness from Adversarial Human Attack](https://mlsanity.com/paper/1908.06083/) (2019)
- [German's Next Language Model](https://mlsanity.com/paper/2010.10906/) (2020)
- [MultiVerS: Improving scientific claim verification with weak supervision and full-document context](https://mlsanity.com/paper/2112.01640/) (2021)
- [Directions in Abusive Language Training Data: Garbage In, Garbage Out](https://mlsanity.com/paper/2004.01670/) (2020)
- [Beat the AI: Investigating Adversarial Human Annotation for Reading Comprehension](https://mlsanity.com/paper/2002.00293/) (2020)
- [Improving Question Answering Model Robustness with Synthetic Adversarial Data Generation](https://mlsanity.com/paper/2104.08678/) (2021)
- [ETHOS: an Online Hate Speech Detection Dataset](https://mlsanity.com/paper/2006.08328/) (2020)
- [Multilingual HateCheck: Functional Tests for Multilingual Hate Speech Detection Models](https://mlsanity.com/paper/2206.09917/) (2022)
- [Hatemoji: A Test Suite and Adversarially-Generated Dataset for Benchmarking and Detecting Emoji-based Hate](https://mlsanity.com/paper/2108.05921/) (2021)
- ["Call me sexist, but...": Revisiting Sexism Detection Using Psychological Scales and Adversarial Samples](https://mlsanity.com/paper/2004.12764/) (2020)
- [Adversarially Regularising Neural NLI Models to Integrate Logical Background Knowledge](https://mlsanity.com/paper/1808.08609/) (2018)
- [Models in the Loop: Aiding Crowdworkers with Generative Annotation Assistants](https://mlsanity.com/paper/2112.09062/) (2021)
- [People Make Better Edits: Measuring the Efficacy of LLM-Generated Counterfactually Augmented Data for Harmful Language Detection](https://mlsanity.com/paper/2311.01270/) (2023)
- [Analyzing Dynamic Adversarial Training Data in the Limit](https://mlsanity.com/paper/2110.08514/) (2021)
- [Hate Speech Criteria: A Modular Approach to Task-Specific Hate Speech Definitions](https://mlsanity.com/paper/2206.15455/) (2022)
- [Mastering the Dungeon: Grounded Language Learning by Mechanical Turker Descent](https://mlsanity.com/paper/1711.07950/) (2017)
- [Are Chess Discussions Racist? An Adversarial Hate Speech Data Set](https://mlsanity.com/paper/2011.10280/) (2020)
- [Adversarial Attacks and Defenses for Social Network Text Processing Applications: Techniques, Challenges and Future Research Directions](https://mlsanity.com/paper/2110.13980/) (2021)

## Built on by (most cited first, 13 of 13)

- [From Languages to Geographies: Towards Evaluating Cultural Bias in Hate Speech Datasets](https://mlsanity.com/paper/2404.17874/) (2024)
- [HateDay: Insights from a Global Hate Speech Dataset Representative of a Day on Twitter](https://mlsanity.com/paper/2411.15462/) (2024)
- [Can Prompting LLMs Unlock Hate Speech Detection across Languages? A Zero-shot and Few-shot Study](https://mlsanity.com/paper/2505.06149/) (2025)
- [A Modular Taxonomy for Hate Speech Definitions and Its Impact on Zero-Shot LLM Classification Performance](https://mlsanity.com/paper/2506.18576/) (2025)
- [Unified Game Moderation: Soft-Prompting and LLM-Assisted Label Transfer for Resource-Efficient Toxicity Detection](https://mlsanity.com/paper/2506.06347/) (2025)
- [AustroTox: A Dataset for Target-Based Austrian German Offensive Language Detection](https://mlsanity.com/paper/2406.08080/) (2024)
- [Data-Efficient Hate Speech Detection via Cross-Lingual Nearest Neighbor Retrieval with Limited Labeled Data](https://mlsanity.com/paper/2505.14272/) (2025)
- [Mapping Toxic Comments Across Demographics: A Dataset from German Public Broadcasting](https://mlsanity.com/paper/2508.21084/) (2025)
- [Bye Bye Perspective API: Lessons for Building and Governing Measurement Infrastructure](https://mlsanity.com/paper/2604.25580/) (2026)
- [Web(er) of Hate: A Survey on How Hate Speech Is Typed](https://mlsanity.com/paper/2506.16190/) (2025)
- [Leveraging the Potential of Prompt Engineering for Hate Speech Detection in Low-Resource Languages](https://mlsanity.com/paper/2506.23930/) (2025)
- [Toward Generalized Cross-Lingual Hateful Language Detection with Web-Scale Data and Ensemble LLM Annotations](https://mlsanity.com/paper/2604.09625/) (2026)
- [MUCnoHARM@GermEval Shared Task 2026: Retrieval-based In-Context Learning for Defamatory Offences, and Where It Falls Short](https://mlsanity.com/paper/2609.09791/) (2026)

## Abstract (arXiv)

Hate speech detection models are only as good as the data they are trained on. Datasets sourced from social media suffer from systematic gaps and biases, leading to unreliable models with simplistic decision boundaries. Adversarial datasets, collected by exploiting model weaknesses, promise to fix this problem. However, adversarial data collection can be slow and costly, and individual annotators have limited creativity. In this paper, we introduce GAHD, a new German Adversarial Hate speech Dataset comprising ca. 11k examples. During data collection, we explore new strategies for supporting annotators, to create more diverse adversarial examples more efficiently and provide a manual analysis of annotator disagreements for each strategy. Our experiments show that the resulting dataset is challenging even for state-of-the-art hate speech detection models, and that training on GAHD clearly improves model robustness. Further, we find that mixing multiple support strategies is most advantageous. We make GAHD publicly available at https://github.com/jagol/gahd.