HERB: Measuring Hierarchical Regional Bias in Pre-trained Language Models

TL;DR

This paper introduces HERB, a hierarchical evaluation method to measure regional bias in pre-trained language models, revealing biases influenced by geographical clustering and affecting downstream NLP tasks.

Contribution

The paper presents a novel hierarchical metric, HERB, for quantifying regional bias in language models based on sub-region clustering, addressing a previously unexplored aspect of bias.

Findings

Regional biases are influenced by geographical clustering.

HERB effectively evaluates regional bias across topics.

Biases can propagate to downstream NLP tasks.

Abstract

Fairness has become a trending topic in natural language processing (NLP), which addresses biases targeting certain social groups such as genders and religions. However, regional bias in language models (LMs), a long-standing global discrimination problem, still remains unexplored. This paper bridges the gap by analysing the regional bias learned by the pre-trained language models that are broadly used in NLP tasks. In addition to verifying the existence of regional bias in LMs, we find that the biases on regional groups can be strongly influenced by the geographical clustering of the groups. We accordingly propose a HiErarchical Regional Bias evaluation method (HERB) utilising the information from the sub-region clusters to quantify the bias in pre-trained LMs. Experiments show that our hierarchical metric can effectively evaluate the regional bias with respect to comprehensive topics and measure the potential regional bias that can be propagated to downstream tasks. Our codes are available at https://github.com/Bernard-Yang/HERB.