Word embeddings derived from human-generated corpora inherit strong gender
bias which can be further amplified by downstream models. Some commonly adopted
debiasing approaches, including the seminal Hard Debias algorithm, apply
post-processing procedures that project pre-trained word embeddings into a
subspace orthogonal to an inferred gender subspace. We discover that
semantic-agnostic corpus regularities such as word frequency captured by the
word embeddings negatively impact the performance of these algorithms. We
propose a simple but effective technique, Double Hard Debias, which purifies
the word embeddings against such corpus regularities prior to inferring and
removing the gender subspace. Experiments on three bias mitigation benchmarks
show that our approach preserves the distributional semantics of the
pre-trained word embeddings while reducing gender bias to a significantly
larger degree than prior approaches.

本文研究了人类生成语料库（human-generated corpora）中由来的单词嵌入（word embeddings）存在着的性别偏见问题，以及已有的性别去偏置（debiasing）方法的局限性；在此基础上，提出了一种新的技术，Double Hard Debias，该技术能够在推断并去除性别子空间之前，首先过滤掉语料库中存储的单词使用频率等无关因素。实验证明，与之前的方法相比，本文提出的技术不仅保留了预训练好的单词嵌入的分布语义（distributional semantics），而且能够更有效地降低性别偏见。