Us-vs-Them bias in Large Language Models

TL;DR

This paper explores social identity bias in large language models, revealing how personas influence ingroup/outgroup attitudes and demonstrating a mitigation method to reduce bias.

Contribution

It provides a comprehensive analysis of social bias in LLMs and introduces ION, a fine-tuning approach to mitigate 'us versus them' bias.

Findings

Consistent ingroup-positive and outgroup-negative associations across models

Persona conditioning alters linguistic and semantic patterns

Bias mitigation reduces sentiment divergence by up to 69%

Abstract

This study investigates ``us versus them'' bias, as described by Social Identity Theory, in large language models (LLMs) under both default and persona-conditioned settings across multiple architectures (GPT-4.1, DeepSeek-3.1, Gemma-2.0, Grok-3.0, and LLaMA-3.1). Using sentiment dynamics, allotaxonometry, and embedding regression, we find consistent ingroup-positive and outgroup-negative associations across foundational LLMs. We find that adopting a persona systematically alters models' evaluative and affiliative language patterns. For the exemplar personas examined, conservative personas exhibit greater outgroup hostility, whereas liberal personas display stronger ingroup solidarity. Persona conditioning produces distinct clustering in embedding space and measurable semantic divergence, supporting the view that even abstract identity cues can shift models' linguistic behavior. Furthermore, outgroup-targeted prompts increased hostility bias by 1.19--21.76\% across models. These findings suggest that LLMs learn not only factual associations about social groups but also internalize and reproduce distinct ways of being, including attitudes, worldviews, and cognitive styles that are activated when enacting personas. We interpret these results as evidence of a multi-scale coupling between local context (e.g., the persona prompt), localizable representations (what the model ``knows''), and global cognitive tendencies (how it ``thinks''), which are at least reflected in the training data. Finally, we demonstrate ION, an ``us versus them'' bias mitigation approach using fine-tuning and direct preference optimization (DPO), which reduces sentiment divergence by up to 69\%, highlighting the potential for targeted mitigation strategies in future LLM development.