# Beyond Detection: Unveiling Fairness Vulnerabilities in Abusive Language Models

arXiv: https://arxiv.org/abs/2311.09428 · HTML: https://mlsanity.com/paper/2311.09428/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 44 references that are in the map.
- First posted on arXiv 15 Nov 2023.

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

- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Equality of Opportunity in Supervised Learning](https://mlsanity.com/paper/1610.02413/) (2016)
- [Automated Hate Speech Detection and the Problem of Offensive Language](https://mlsanity.com/paper/1703.04009/) (2017)
- [Evasion Attacks against Machine Learning at Test Time](https://mlsanity.com/paper/1708.06131/) (2017)
- [Mitigating Unwanted Biases with Adversarial Learning](https://mlsanity.com/paper/1801.07593/) (2018)
- [Poison Frogs! Targeted Clean-Label Poisoning Attacks on Neural Networks](https://mlsanity.com/paper/1804.00792/) (2018)
- [Weight Poisoning Attacks on Pre-trained Models](https://mlsanity.com/paper/2004.06660/) (2020)
- [Manipulating Machine Learning: Poisoning Attacks and Countermeasures for Regression Learning](https://mlsanity.com/paper/1804.00308/) (2018)
- [The Variational Fair Autoencoder](https://mlsanity.com/paper/1511.00830/) (2015)
- [Racial Bias in Hate Speech and Abusive Language Detection Datasets](https://mlsanity.com/paper/1905.12516/) (2019)
- [Mind the Style of Text! Adversarial and Backdoor Attacks Based on Text Style Transfer](https://mlsanity.com/paper/2110.07139/) (2021)
- [Reducing Gender Bias in Abusive Language Detection](https://mlsanity.com/paper/1808.07231/) (2018)
- [Stronger Data Poisoning Attacks Break Data Sanitization Defenses](https://mlsanity.com/paper/1811.00741/) (2018)
- [Backdoor Attacks and Countermeasures on Deep Learning: A Comprehensive Review](https://mlsanity.com/paper/2007.10760/) (2020)
- [Deceiving Google's Perspective API Built for Detecting Toxic Comments](https://mlsanity.com/paper/1702.08138/) (2017)
- [Technical Report: When Does Machine Learning FAIL? Generalized Transferability for Evasion and Poisoning Attacks](https://mlsanity.com/paper/1803.06975/) (2018)
- [Demoting Racial Bias in Hate Speech Detection](https://mlsanity.com/paper/2005.12246/) (2020)
- [All You Need is "Love": Evading Hate-speech Detection](https://mlsanity.com/paper/1808.09115/) (2018)
- [Demographics Should Not Be the Reason of Toxicity: Mitigating Discrimination in Text Classifications with Instance Weighting](https://mlsanity.com/paper/2004.14088/) (2020)
- [Stereotypical Bias Removal for Hate Speech Detection Task using Knowledge-based Generalizations](https://mlsanity.com/paper/2001.05495/) (2020)
- [Adversarial Deep Ensemble: Evasion Attacks and Defenses for Malware Detection](https://mlsanity.com/paper/2006.16545/) (2020)
- [Natural Backdoor Attack on Text Data](https://mlsanity.com/paper/2006.16176/) (2020)
- [Poisoning Attacks on Fair Machine Learning](https://mlsanity.com/paper/2110.08932/) (2021)
- [Towards Fair Classification against Poisoning Attacks](https://mlsanity.com/paper/2210.09503/) (2022)

## Built on by (most cited first, 3 of 3)

- [Sora: A Review on Background, Technology, Limitations, and Opportunities of Large Vision Models](https://mlsanity.com/paper/2402.17177/) (2024)
- [Atlas: A Framework for ML Lifecycle Provenance & Transparency](https://mlsanity.com/paper/2502.19567/) (2025)
- [Injecting Bias into Text Classification Models using Backdoor Attacks](https://mlsanity.com/paper/2412.18975/) (2024)

## Abstract (arXiv)

This work investigates the potential of undermining both fairness and detection performance in abusive language detection. In a dynamic and complex digital world, it is crucial to investigate the vulnerabilities of these detection models to adversarial fairness attacks to improve their fairness robustness. We propose a simple yet effective framework FABLE that leverages backdoor attacks as they allow targeted control over the fairness and detection performance. FABLE explores three types of trigger designs (i.e., rare, artificial, and natural triggers) and novel sampling strategies. Specifically, the adversary can inject triggers into samples in the minority group with the favored outcome (i.e., "non-abusive") and flip their labels to the unfavored outcome, i.e., "abusive". Experiments on benchmark datasets demonstrate the effectiveness of FABLE attacking fairness and utility in abusive language detection.