Speciesist Language and Nonhuman Animal Bias in English Masked Language Models

TL;DR

This paper investigates speciesist bias in English Masked Language Models, revealing that models tend to associate harmful language with nonhuman animals, highlighting a previously overlooked social bias in NLP.

Contribution

It introduces the first analysis of speciesist bias in NLP models, specifically in BERT, and provides experimental evidence of such biases against various animal names.

Findings

Models associate harmful words with nonhuman animals.

Bias varies across different animal names.

Code for experiments will be publicly available.

Abstract

Various existing studies have analyzed what social biases are inherited by NLP models. These biases may directly or indirectly harm people, therefore previous studies have focused only on human attributes. However, until recently no research on social biases in NLP regarding nonhumans existed. In this paper, we analyze biases to nonhuman animals, i.e. speciesist bias, inherent in English Masked Language Models such as BERT. We analyzed speciesist bias against 46 animal names using template-based and corpus-extracted sentences containing speciesist (or non-speciesist) language. We found that pre-trained masked language models tend to associate harmful words with nonhuman animals and have a bias toward using speciesist language for some nonhuman animal names. Our code for reproducing the experiments will be made available on GitHub.