From Bias to Positionality: A critical grammar for Large Language Models
cs.CL, cs.AI, cs.HC
Submitted: 2026-11-10
Updated: 2026-11-10
Code: https://github.com/jfisher52/Approximation_Political_Neutrality
Terminology
Sources
- Corporations Constitute Intelligence
- Operationalizing Pluralistic Values in Large Language Model Alignment Reveals Trade-offs in Safety, Inclusivity, and Model Behavior
- Constitutional AI: Harmlessness from AI Feedback
- The Values Encoded in Machine Learning Research
- Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings
- Auditing Alignment Controllability in LLMs via Political Axes
- Large Language Models Reflect the Ideology of their Creators
- What Is The Political Content in LLMs' Pre- and Post-Training Data?
- Persona Vectors: Monitoring and Controlling Character Traits in Language Models
- Dated Data: Tracing Knowledge Cutoffs in Large Language Models
- Believing Anthropomorphism: Examining the Role of Anthropomorphic Cues on Trust in Large Language Models
- Neural Retrievers are Biased Towards LLM-Generated Content
- A Broader View on Bias in Automated Decision-Making: Reflecting on Epistemology and Dynamics
- Political Neutrality in AI Is Impossible- But Here Is How to Approximate It
- Disclosure By Design: Identity Transparency as a Behavioural Property of Conversational AI Models
- Datasheets for Datasets
- What if AI systems weren't chatbots?
- Values in the Wild: Discovering and Analyzing Values in Real-World Language Model Interactions
- How Well Do Models Follow Their Constitutions?
- Randomness, Not Representation: The Unreliability of Evaluating Cultural Alignment in LLMs
Related papers
- ALARM: Audio-Language Alignment for Reasoning Models
- ThinkSum: Probabilistic reasoning over sets using large language models
- Coherence boosting: When your pretrained language model is not paying enough attention
- From Facts to Insights: A Persona-Driven Dual Memory Framework and Dataset for Role-Playing Agents
- Improving General Role-Playing Agents via Psychology-Grounded Reasoning and Role-Aware Policy Optimization
- The Art of Asking: Multilingual Prompt Optimization for Synthetic Data