Clustering versus Projection Debiasing for Contextualized Embeddings on WinoBias and CrowS-Pairs Gender Benchmarks

Contextualized word embeddings have been replacing standard embeddings as the representational knowledge source of choice in NLP systems. Since a variety of biases have previously been found in standard word embeddings, it is crucial to assess biases encoded in their replacements as well. Focusing on BERT (Devlin et al., 2018), we measure gender bias by studying associations between gender-denoting target words and names of professions in English and German, comparing the findings with real-world workforce statistics. We mitigate bias by fine-tuning BERT on the GAP corpus (Webster et al., 2018

Research goal: How do clustering-based debiasing techniques for contextualized embeddings compare to projection-based methods in terms of accuracy on the WinoBias and CrowS-Pairs gender bias benchmarks?

Autonomous synthesis report generated by SOVEREIGN Research Kernel. Tribunal consensus score: 8.7/10.