# Understanding and Mitigating Political Stance Cross-topic Generalization in Large Language Models

arXiv: https://arxiv.org/abs/2508.02360 · HTML: https://mlsanity.com/paper/2508.02360/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 54 references that are in the map.
- First posted on arXiv 4 Aug 2025.

## Builds on (most cited first, 27 of 27)

- [Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 small](https://mlsanity.com/paper/2211.00593/) (2022)
- [Whose Opinions Do Language Models Reflect?](https://mlsanity.com/paper/2303.17548/) (2023)
- [Knowledge Neurons in Pretrained Transformers](https://mlsanity.com/paper/2104.08696/) (2021)
- [Towards Best Practices of Activation Patching in Language Models: Metrics and Methods](https://mlsanity.com/paper/2309.16042/) (2023)
- [Finding Neurons in a Haystack: Case Studies with Sparse Probing](https://mlsanity.com/paper/2305.01610/) (2023)
- [DetectLLM: Leveraging Log Rank Information for Zero-Shot Detection of Machine-Generated Text](https://mlsanity.com/paper/2306.05540/) (2023)
- [Political Compass or Spinning Arrow? Towards More Meaningful Evaluations for Values and Opinions in Large Language Models](https://mlsanity.com/paper/2402.16786/) (2024)
- [Backdooring Instruction-Tuned Large Language Models with Virtual Prompt Injection](https://mlsanity.com/paper/2307.16888/) (2023)
- [The Political Preferences of LLMs](https://mlsanity.com/paper/2402.01789/) (2024)
- [Are Large Language Models Consistent over Value-laden Questions?](https://mlsanity.com/paper/2407.02996/) (2024)
- [Confidence Regulation Neurons in Language Models](https://mlsanity.com/paper/2406.16254/) (2024)
- [Large Language Models Reflect the Ideology of their Creators](https://mlsanity.com/paper/2410.18417/) (2024)
- [Understanding Aha Moments: from External Observations to Internal Mechanisms](https://mlsanity.com/paper/2504.02956/) (2025)
- [Locate-then-edit for Multi-hop Factual Recall under Knowledge Editing](https://mlsanity.com/paper/2410.06331/) (2024)
- [Understanding the Repeat Curse in Large Language Models from a Feature Perspective](https://mlsanity.com/paper/2504.14218/) (2025)
- [Who Wrote This? The Key to Zero-Shot LLM-Generated Text Detection Is GECScore](https://mlsanity.com/paper/2405.04286/) (2024)
- [Learnable Privacy Neurons Localization in Language Models](https://mlsanity.com/paper/2405.10989/) (2024)
- [How Susceptible are Large Language Models to Ideological Manipulation?](https://mlsanity.com/paper/2402.11725/) (2024)
- [Towards Understanding Multi-Task Learning (Generalization) of LLMs via Detecting and Exploring Task-Specific Neurons](https://mlsanity.com/paper/2407.06488/) (2024)
- [EAP-GP: Mitigating Saturation Effect in Gradient-based Automated Circuit Identification](https://mlsanity.com/paper/2502.06852/) (2025)
- [Is Your LLM-Based Multi-Agent a Reliable Real-World Planner? Exploring Fraud Detection in Travel Planning](https://mlsanity.com/paper/2505.16557/) (2025)
- [Understanding and Mitigating Gender Bias in LLMs via Interpretable Neuron Editing](https://mlsanity.com/paper/2501.14457/) (2025)
- [Understanding and Mitigating Cross-lingual Privacy Leakage via Language-specific and Universal Privacy Neurons](https://mlsanity.com/paper/2506.00759/) (2025)
- [Whose Side Are You On? Investigating the Political Stance of Large Language Models](https://mlsanity.com/paper/2403.13840/) (2024)
- [Linguistic and Embedding-Based Profiling of Texts generated by Humans and Large Language Models](https://mlsanity.com/paper/2507.13614/) (2025)
- [From General to Targeted Rewards: Surpassing GPT-4 in Open-Ended Long-Context Generation](https://mlsanity.com/paper/2506.16024/) (2025)
- [Unveiling Instruction-Specific Neurons & Experts: An Analytical Framework for LLM's Instruction-Following Capabilities](https://mlsanity.com/paper/2505.21191/) (2025)

## Built on by (most cited first, 6 of 6)

- [Adaptive Multi-Subspace Representation Steering for Attribute Alignment in Large Language Models](https://mlsanity.com/paper/2508.10599/) (2025)
- [Beyond Scalars: Evaluating and Understanding LLM Reasoning via Geometric Progress and Stability](https://mlsanity.com/paper/2603.10384/) (2026)
- [Understanding the Dynamics of Demonstration Conflict in In-Context Learning](https://mlsanity.com/paper/2603.04464/) (2026)
- [Word Recovery in Large Language Models Enables Character-Level Tokenization Robustness](https://mlsanity.com/paper/2603.10771/) (2026)
- [Innocuous-Seeming Data, Latent Ideology: Ideological Generalisation in Finetuned LLMs](https://mlsanity.com/paper/2607.14888/) (2026)
- [Multi-Adapter Representation Interventions via Energy Calibration](https://mlsanity.com/paper/2605.28722/) (2026)

## Abstract (arXiv)

Fine-tuning Large Language Models on a political topic will significantly manipulate their political stance on various issues and unintentionally affect their stance on unrelated topics. While previous studies have proposed this issue, there is still a lack of understanding regarding the internal representations of these stances and the mechanisms that lead to unintended cross-topic generalization. In this paper, we systematically explore the internal mechanisms underlying this phenomenon from a neuron-level perspective and how to mitigate the cross-topic generalization of political fine-tuning. Firstly, we propose Political Neuron Localization through Activation Contrasting (PNLAC) to identify two distinct types of political neurons: general political neurons, which govern stance across multiple political topics, and topic-specific neurons} that affect the model's political stance on individual topics. We find the existence of these political neuron types across four models and datasets through activation patching experiments. Leveraging these insights, we introduce InhibitFT, an inhibition-based fine-tuning method, effectively mitigating the cross-topic stance generalization. Experimental results demonstrate the robustness of identified neuron types across various models and datasets, and show that InhibitFT significantly reduces the cross-topic stance generalization by 20% on average, while preserving topic-specific performance. Moreover, we demonstrate that selectively inhibiting only 5% of neurons is sufficient to effectively mitigate the cross-topic stance generalization.