Quantifying and alleviating political bias in language models

2022/01/03 by Ruibo Liu, Chenyan Jia, Jason Wei +2 · 18 citations
Computer Science · #Topic Modeling #Explainable Artificial Intelligence (XAI) #Text Readability and Simplification

paper · doi:10.1016/j.artint.2021.103654

Cited by

• On the inevitability of left-leaning political bias in aligned language models
• Understanding Down Syndrome Stereotypes in LLM-Based Personas
• Can Finetuing LLMs on Small Human Samples Increase Heterogeneity, Alignment, and Belief-Action Coherence?
• Prioritize Economy or Climate Action? Investigating ChatGPT Response Differences Based on Inferred Political Orientation
• Hijacking algorithmic bias: analyzing the political discourse around ChatGPT on social media
• The end of experimental research as we know it? A perspective on generative artificial intelligence in communication science
• Prompt selection matters: enhancing text annotations for social sciences with large language models
• Assessing political bias in AI systems: a framework for disentangling viewpoint preferences from epistemic integrity
• Simulating a Bias Mitigation Scenario in Large Language Models
• Prompt Commons: Collective Prompting as Governance for Urban AI
• Should LLMs be WEIRD? Exploring WEIRDness and Human Rights in Large Language Models
• Attention-Enabled Ensemble Deep Learning Models and Their Validation for Depression Detection: A Domain Adoption Paradigm. [europepmc]
• Empathy and Equity: Key Considerations for Large Language Model Adoption in Health Care. [europepmc]
• Exploring the role of AI in classifying, analyzing, and generating case reports on assisted suicide cases: feasibility and ethical implications. [europepmc]
• The political preferences of LLMs. [europepmc]
• Detecting implicit biases of large language models with Bayesian hypothesis testing. [europepmc]
• LLMs Can Fuel Extremist Attitudes Using Universal Moral Framings. [europepmc]
• An AI-powered research assistant in the lab: A practical guide for text analysis through iterative collaboration with LLMs. [europepmc]