What is in a name? Mitigating Name Bias in Text Embeddings via Anonymization

TL;DR

This paper identifies and mitigates name bias in text-embeddings by proposing a text anonymization method that removes names during inference, improving the accuracy of semantic similarity assessments.

Contribution

It introduces a simple, training-free anonymization technique to reduce name bias in text-embedding models, enhancing downstream NLP task performance.

Findings

Name bias exists in various text-embedding models.

Anonymization improves semantic similarity accuracy.

Method is practical and easy to implement.

Abstract

Text-embedding models often exhibit biases arising from the data on which they are trained. In this paper, we examine a hitherto unexplored bias in text-embeddings: bias arising from the presence of $\textit{names}$ such as persons, locations, organizations etc. in the text. Our study shows how the presence of $\textit{name-bias}$ in text-embedding models can potentially lead to erroneous conclusions in assessment of thematic similarity.Text-embeddings can mistakenly indicate similarity between texts based on names in the text, even when their actual semantic content has no similarity or indicate dissimilarity simply because of the names in the text even when the texts match semantically. We first demonstrate the presence of name bias in different text-embedding models and then propose $\textit{text-anonymization}$ during inference which involves removing references to names, while preserving the core theme of the text. The efficacy of the anonymization approach is demonstrated on two downstream NLP tasks, achieving significant performance gains. Our simple and training-optimization-free approach offers a practical and easily implementable solution to mitigate name bias.