Fine-Tuned LLMs are "Time Capsules" for Tracking Societal Bias Through Books

TL;DR

This paper introduces BookPAGE, a novel method using fine-tuned LLMs on decade-specific books to track societal bias evolution over time, revealing how biases reflect historical cultural shifts.

Contribution

We develop a new corpus and methodology to analyze societal biases in LLMs based on literary content, highlighting bias trends aligned with historical societal changes.

Findings

Biases in LLMs mirror societal changes over decades.

Portrayal of women in leadership roles increased over time.

References to LGBTQ+ relationships grew significantly from 1980s to 2000s.

Abstract

Books, while often rich in cultural insights, can also mirror societal biases of their eras - biases that Large Language Models (LLMs) may learn and perpetuate during training. We introduce a novel method to trace and quantify these biases using fine-tuned LLMs. We develop BookPAGE, a corpus comprising 593 fictional books across seven decades (1950-2019), to track bias evolution. By fine-tuning LLMs on books from each decade and using targeted prompts, we examine shifts in biases related to gender, sexual orientation, race, and religion. Our findings indicate that LLMs trained on decade-specific books manifest biases reflective of their times, with both gradual trends and notable shifts. For example, model responses showed a progressive increase in the portrayal of women in leadership roles (from 8% to 22%) from the 1950s to 2010s, with a significant uptick in the 1990s (from 4% to 12%), possibly aligning with third-wave feminism. Same-sex relationship references increased markedly from the 1980s to 2000s (from 0% to 10%), mirroring growing LGBTQ+ visibility. Concerningly, negative portrayals of Islam rose sharply in the 2000s (26% to 38%), likely reflecting post-9/11 sentiments. Importantly, we demonstrate that these biases stem mainly from the books' content and not the models' architecture or initial training. Our study offers a new perspective on societal bias trends by bridging AI, literary studies, and social science research.