Lune

EMNLP Findings2025

Probing Political Ideology in Large Language Models: How Latent Political Representations Generalize Across Tasks

Tianyi Zhang

Abstract

Large language models (LLMs) encode rich internal representations of political ideology, but it remains unclear how these representations contribute to model decision-making, and how these latent dimensions interact with one another. We apply inference-time interventions on ideological directions identified via linear probes to steer LLMs along learned ideological directions, and evaluate their effect on three tasks: political bias detection, voting preference simulation, and bias neutralization. Our results show that learned ideological representations generalize well to bias detection, but not as well to voting simulations, suggesting that political ideology is encoded in multiple, partially disentangled latent structures. We also observe asymmetries in how interventions affect liberal versus conservative outputs and across models, raising concerns about pretraininginduced bias and post-training alignment effects. This work highlights the risks of using biased LLMs for politically sensitive tasks, and calls for deeper investigation into the interaction of social dimensions in model representations, as well as methods for steering them toward fairer, more transparent behavior 1 .

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

Questions to start from

Lune can draw on

the full paper · related work

Get the answer

Free to start. No credit card required.

Lineage

  • The Linear Representation Hypothesis and the Geometry of Large Language Models cites Language Models Represent Space and Time
Builds onCited byLeft to right by year; the more cited, the higher.

Builds on11

  1. Inference-Time Intervention: Eliciting Truthful Answers from a Language ModelKenneth Li, Oam Patel, Fernanda B. Viégas, Hanspeter Pfister et al.NeurIPS 2023 · 1,549 citations
  2. Whose Opinions Do Language Models Reflect?Shibani Santurkar, Esin Durmus, Faisal Ladhak, Cinoo Lee et al.ICML 2023 · 764 citations
  3. The Linear Representation Hypothesis and the Geometry of Large Language ModelsKiho Park, Yo Joong Choe, Victor VeitchICML 2024 · 461 citations
  4. Language Models Represent Space and TimeWes Gurnee, Max TegmarkICLR 2024 · 303 citations
  5. Bias Runs Deep: Implicit Reasoning Biases in Persona-Assigned LLMsShashank Gupta, Vaishnavi Shrivastava, Ameet Deshpande, Ashwin Kalyan et al.ICLR 2024 · 212 citations

Related papers

  1. Linear Representations of Political Perspective Emerge in Large Language ModelsJunsol Kim, James Evans, Aaron ScheinICLR 2025
  2. How Susceptible are Large Language Models to Ideological Manipulation?Kai Chen, Zihao He, Jun Yan, Taiwei Shi et al.EMNLP 2024 · 4 citations
  3. Whose Emotions and Moral Sentiments do Language Models Reflect?Zihao He, Siyi Guo, Ashwin Rao, Kristina LermanACL Findings 2024 · 5 citations
  4. Reading Between the Tokens: Improving Preference Predictions through Mechanistic ForecastingSarah Ball, Simeon Allmendinger, Frauke Kreuter, Niklas KühlICML 2026
  5. Examining Alignment of Large Language Models through Representative Heuristics: the case of political stereotypesSullam Jeoung, Yubin Ge, Haohan Wang, Jana DiesnerICLR 2025
Dusk over the sea between two cliffs drawn in fine vertical lines