mlsanityData to 7 Oct 2026

Ideology as a Problem: Lightweight Logit Steering for Annotator-Specific Alignment in Social Media Analysis

arXiv 2601.04207 ↗First posted 8 Dec 2025cs.CL cs.AI cs.SIcited by 0

Not on the map. Why?

Abstract (arXiv)

LLMs internally organize political ideology along low-dimensional structures that are partially, but not fully aligned with human ideological space. This misalignment is systematic, model specific, and measurable. We introduce a lightweight linear probe that both quantifies the misalignment and minimally corrects the output layer. This paper introduces a simple and efficient method for aligning models with specific user opinions. Instead of retraining the model, we calculated a bias score from its internal features and directly adjusted the final output probabilities. This solution is practical and low-cost and preserves the original reasoning power of the model.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 25 of its 28 references that are in the map. First posted on arXiv 8 Dec 2025.

Builds on

25 of 28 references.

References · 25

  1. 2022Training language models to follow instructions with human feedbackcited by 15,712
  2. 2024The Llama 3 Herd of Modelscited by 14,192
  3. 2025Qwen3 Technical Reportcited by 8,195
  4. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 7,916
  5. 2016Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddingscited by 1,304
  6. 2023Representation Engineering: A Top-Down Approach to AI Transparencycited by 1,239
  7. 2023Steering Language Models With Activation Engineeringcited by 824
  8. 2023Whose Opinions Do Language Models Reflect?cited by 759
  9. 2022Out of One, Many: Using Language Models to Simulate Human Samplescited by 737
  10. 2022Discovering Latent Knowledge in Language Models Without Supervisioncited by 733
  11. 2023The Linear Representation Hypothesis and the Geometry of Large Language Modelscited by 679
  12. 2023Towards Measuring the Representation of Subjective Global Opinions in Language Modelscited by 424
13 more
  1. 2023DoLa: Decoding by Contrasting Layers Improves Factuality in Large Language Modelscited by 367
  2. 2021Dealing with Disagreements: Looking Beyond the Majority Vote in Subjective Annotationscited by 306
  3. 2023From Pretraining Data to Language Models to Downstream Tasks: Tracking the Trails of Political Biases Leading to Unfair NLP Modelscited by 257
  4. 2022Fine-tuning language models to find agreement among humans with diverse preferencescited by 189
  5. 2023The political ideology of conversational AI: Converging evidence on ChatGPT's pro-environmental, left-libertarian orientationcited by 183
  6. 2023Do LLMs exhibit human-like response biases? A case study in survey designcited by 176
  7. 2021All Bark and No Bite: Rogue Dimensions in Transformer Language Models Obscure Representational Qualitycited by 150
  8. 2022Jury Learning: Integrating Dissenting Voices into Machine Learning Modelscited by 115
  9. 2023Value Kaleidoscope: Engaging AI with Pluralistic Human Values, Rights, and Dutiescited by 97
  10. 2018The Geometry of Culture: Analyzing Meaning through Word Embeddingscited by 87
  11. 2024The Political Preferences of LLMscited by 83
  12. 2020Measuring a hate speech spectrum with faceted Rasch item response theory and perspective-aware, explainable-by-design deep learningcited by 67
  13. 2020Emergence of Separable Manifolds in Deep Language Representationscited by 39

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Political Alignment in Large Language Models: A Multidimensional Audit of Psychometric Identity and Behavioral Bias
  2. 2025PReSS: An Automated Black-Box Framework for Evaluating Political Stance Stability in LLMs
  3. 2024GermanPartiesQA: Benchmarking Commercial Large Language Models and AI Companions for Political Alignment and Sycophancy
  4. 2026LLMs Can Infer Political Alignment from Online Conversations
  5. 2025Examining Alignment of Large Language Models through Representative Heuristics: The Case of Political Stereotypes
  6. 2025Language, Culture, and Ideology: Personalizing Offensiveness Detection in Political Tweets with Reasoning LLMs
  7. 2025Confident, Calibrated, or Complicit: Safety Alignment and Ideological Bias in LLM Hate Speech Detection
  8. 2024COMMUNITY-CROSS-INSTRUCT: Unsupervised Instruction Generation for Aligning Large Language Models to Online Communities