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

ACL ARR 2026 January Submission2721 Authors

Keywords: Large Language Models, Mechanistic Interpretability, Political Stance

Abstract: Fine-tuning Large Language Models on a political topic will significantly manipulate their political stance on various issues and unintentionally affect their stance on broad 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 that these political neuron types exist in the middle and later layers across four models and datasets through activation patching experiments. Leveraging these insights, we introduce InhibitFT, an inhibition-based fine-tuning method that effectively mitigates the cross-topic stance generalization. Experimental results demonstrate the robustness of the 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.

Paper Type: Long

Research Area: Interpretability and Analysis of Models for NLP

Research Area Keywords: Language/cultural bias analysis, Interpretability, NLP tools for social analysis

Contribution Types: Model analysis & interpretability

Languages Studied: English

Submission Number: 2721