Intentional Biases in LLM Responses
Not on the map. Why?
Abstract (arXiv)
In this study we intentionally introduce biases into large language model responses in an attempt to create specific personas for interactive media purposes. We explore the differences between open source models such as Falcon-7b and the GPT-4 model from Open AI, and we quantify some differences in responses afforded by the two systems. We find that the guardrails in the GPT-4 mixture of experts models with a supervisor, while useful in assuring AI alignment in general, are detrimental in trying to construct personas with a variety of uncommon viewpoints. This study aims to set the groundwork for future exploration in intentional biases of large language models such that these practices can be applied in the creative field, and new forms of media.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 11 of its 16 references that are in the map. First posted on arXiv 11 Nov 2023.
Builds on
11 of 16 references.
References · 11
- 2023Universal and Transferable Adversarial Attacks on Aligned Language Modelscited by 3,104
- 2023Sparks of Artificial General Intelligence: Early experiments with GPT-4cited by 2,592
- 2021GLaM: Efficient Scaling of Language Models with Mixture-of-Expertscited by 897
- 2023The RefinedWeb Dataset for Falcon LLM: Outperforming Curated Corpora with Web Data, and Web Data Onlycited by 763
- 2023The Internal State of an LLM Knows When It's Lyingcited by 659
- 2023Marked Personas: Using Natural Language Prompts to Measure Stereotypes in Language Modelscited by 207
- 2021Efficient Large Scale Language Modeling with Mixtures of Expertscited by 175
- 2023Personality Traits in Large Language Modelscited by 169
- 2022Ask Me Anything: A simple strategy for prompting language modelscited by 155
- 2023Hallucinations in Large Multilingual Translation Modelscited by 129
- 2023AI Text-to-Behavior: A Study In Steerabilitycited by 11
Built on by
Citing papers per year: 2024: 2.
2 papers.
- 2024Safeguarding Large Language Models: A Surveycited by 78
- 2024Building Guardrails for Large Language Modelscited by 65
- 2024Safeguarding Large Language Models: A Survey3 Jun 2024
- 2024Building Guardrails for Large Language Models2 Feb 2024
Similar, not cited
Close in wording, no citation either way
- 2024Evaluating Large Language Model Biases in Persona-Steered Generation
- 2026From Biased Chatbots to Biased Agents: Examining Role Assignment Effects on LLM Agent Robustness
- 2025Persuasiveness and Bias in LLM: Investigating the Impact of Persuasiveness and Reinforcement of Bias in Language Models
- 2023PersonaLLM: Investigating the Ability of Large Language Models to Express Personality Traits
- 2025Whose Personae? Synthetic Persona Experiments in LLM Research and Pathways to Transparency
- 2024LLM Agents in Interaction: Measuring Personality Consistency and Linguistic Alignment in Interacting Populations of Large Language Models
- 2026Persona Prompting as a Lens on LLM Social Reasoning
- 2025Open Character Training: Shaping the Persona of AI Assistants through Constitutional AI