Challenges in Automated Debiasing for Toxic Language Detection
This addresses fairness and accuracy issues in toxic language detection for AI systems, but it is incremental as it builds on prior debiasing approaches.
The paper tackled the problem of biased associations in toxic language detection, finding that existing debiasing methods are limited and proposing a dialect-aware data correction method that reduces dialectal associations with toxicity.
Biased associations have been a challenge in the development of classifiers for detecting toxic language, hindering both fairness and accuracy. As potential solutions, we investigate recently introduced debiasing methods for text classification datasets and models, as applied to toxic language detection. Our focus is on lexical (e.g., swear words, slurs, identity mentions) and dialectal markers (specifically African American English). Our comprehensive experiments establish that existing methods are limited in their ability to prevent biased behavior in current toxicity detectors. We then propose an automatic, dialect-aware data correction method, as a proof-of-concept. Despite the use of synthetic labels, this method reduces dialectal associations with toxicity. Overall, our findings show that debiasing a model trained on biased toxic language data is not as effective as simply relabeling the data to remove existing biases.