Research map / Large language models / Language, safety and interpretability
FuzzE: Fuzzy Fairness Evaluation of Offensive Language Classifiers on African-American English
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| append offensive words from the source tweet to style-transferred text (lexicon post-processing) (ablation) | FuzzE ensemble · HSOL | Spearman rho FPED: -.578 -> .334 | Helped |
| append offensive words from source (lexicon post-processing) (ablation) | FuzzE ensemble · HSOL | Spearman rho FNED: .323 -> .493 | Helped |
| average fairness estimates over three style transfer methods (ensemble) instead of one method (ablation) | FuzzE · HSOL | Spearman rho avg: .360 -> .413 | Helped |
| ensemble instead of Template alone (ablation) | FuzzE · OLID | Spearman rho avg: .438 -> .400 | Hurt |
| use back-translation instead of retrieval for style transfer (ablation) | FuzzE · OLID | human pass rate: .70 -> .84 | Helped |
| better style transfer method (higher pass rate or fluency) vs better fairness ranking (ablation) | style transfer choice · OLID/HSOL | fairness ranking correlation | Mixed |
Closest papers by meaning
- Spanning the Spectrum of Hatred Detection: A Persian Multi-Label Hate Speech Dataset with Annotator Rationales (AAAI 2024 · similarity 0.83)
- All Should Be Equal in the Eyes of LMs: Counterfactually Aware Fair Text Generation (AAAI 2024 · similarity 0.82)
- A Multi-Task Benchmark for Abusive Language Detection in Low-Resource Settings (NeurIPS 2025 · similarity 0.82)
- Robust Evaluation Measures for Evaluating Social Biases in Masked Language Models (AAAI 2024 · similarity 0.82)
- CEB: Compositional Evaluation Benchmark for Fairness in Large Language Models (ICLR 2025 · similarity 0.81)
- Investigating and Mitigating Undesirable Biases in Large Language Models (AAAI 2025 · similarity 0.81)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.