# Steering Towards Fairness: Mitigating Political Bias in LLMs

arXiv: https://arxiv.org/abs/2508.08846 · HTML: https://mlsanity.com/paper/2508.08846/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 20 references that are in the map.
- First posted on arXiv 12 Aug 2025.

## Builds on (most cited first, 10 of 10)

- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [From Pretraining Data to Language Models to Downstream Tasks: Tracking the Trails of Political Biases Leading to Unfair NLP Models](https://mlsanity.com/paper/2305.08283/) (2023)
- [Large Language Model Alignment: A Survey](https://mlsanity.com/paper/2309.15025/) (2023)
- ["I'm sorry to hear that": Finding New Biases in Language Models with a Holistic Descriptor Dataset](https://mlsanity.com/paper/2205.09209/) (2022)
- [Language Generation Models Can Cause Harm: So What Can We Do About It? An Actionable Survey](https://mlsanity.com/paper/2210.07700/) (2022)
- [Aligning Large Language Models with Human Preferences through Representation Engineering](https://mlsanity.com/paper/2312.15997/) (2023)
- [BERTScore is Unfair: On Social Bias in Language Model-Based Metrics for Text Generation](https://mlsanity.com/paper/2210.07626/) (2022)
- [Analyzing Political Bias and Unfairness in News Articles at Different Levels of Granularity](https://mlsanity.com/paper/2010.10652/) (2020)
- [NeuS: Neutral Multi-News Summarization for Mitigating Framing Bias](https://mlsanity.com/paper/2204.04902/) (2022)
- [Framing Political Bias in Multilingual LLMs Across Pakistani Languages](https://mlsanity.com/paper/2506.00068/) (2025)

## Built on by (most cited first, 7 of 7)

- [Too Helpful, Too Harmless, Too Honest or Just Right?](https://mlsanity.com/paper/2509.08486/) (2025)
- [The Fragility Of Moral Judgment In Large Language Models](https://mlsanity.com/paper/2603.05651/) (2026)
- [Bias Beyond Borders: Political Ideology Evaluation and Steering in Multilingual LLMs](https://mlsanity.com/paper/2601.23001/) (2026)
- [Fairness Evaluation and Inference Level Mitigation in LLMs](https://mlsanity.com/paper/2510.18914/) (2025)
- [Steering Over-refusals Towards Safety in Retrieval Augmented Generation](https://mlsanity.com/paper/2510.10452/) (2025)
- [Do Large Language Models Reflect Demographic Pluralism in Safety?](https://mlsanity.com/paper/2602.07376/) (2026)
- [AuAu: A Benchmark for Auditing Authoritarian Alignment in Large Language Models](https://mlsanity.com/paper/2606.16127/) (2026)

## Abstract (arXiv)

Recent advancements in large language models (LLMs) have enabled their widespread use across diverse real-world applications. However, concerns remain about their tendency to encode and reproduce ideological biases along political and economic dimensions. In this paper, we employ a framework for probing and mitigating such biases in decoder-based LLMs through analysis of internal model representations. Grounded in the Political Compass Test (PCT), this method uses contrastive pairs to extract and compare hidden layer activations from models like Mistral and DeepSeek. We introduce a comprehensive activation extraction pipeline capable of layer-wise analysis across multiple ideological axes, revealing meaningful disparities linked to political framing. Our results show that decoder LLMs systematically encode representational bias across layers, which can be leveraged for effective steering vector-based mitigation. This work provides new insights into how political bias is encoded in LLMs and offers a principled approach to debiasing beyond surface-level output interventions.