mlsanityData to 14 Oct 2026

In-Context Impersonation Reveals Large Language Models' Strengths and Biases

arXiv 2305.14930 ↗First posted 24 May 2023cs.AI cs.CL cs.LGcited by 155

Not on the map. Why?

Abstract (arXiv)

In everyday conversations, humans can take on different roles and adapt their vocabulary to their chosen roles. We explore whether LLMs can take on, that is impersonate, different roles when they generate text in-context. We ask LLMs to assume different personas before solving vision and language tasks. We do this by prefixing the prompt with a persona that is associated either with a social identity or domain expertise. In a multi-armed bandit task, we find that LLMs pretending to be children of different ages recover human-like developmental stages of exploration. In a language-based reasoning task, we find that LLMs impersonating domain experts perform better than LLMs impersonating non-domain experts. Finally, we test whether LLMs' impersonations are complementary to visual information when describing different categories. We find that impersonation can improve performance: an LLM prompted to be a bird expert describes birds better than one prompted to be a car expert. However, impersonation can also uncover LLMs' biases: an LLM prompted to be a man describes cars better than one prompted to be a woman. These findings demonstrate that LLMs are capable of taking on diverse roles and that this in-context impersonation can be used to uncover their hidden strengths and biases.

Key facts. Cited by 155 papers in the map (data through 14 Oct 2026). Builds on 58 of its 97 references that are in the map. First posted on arXiv 24 May 2023.

Builds on

58 of 97 references.

References · 58

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2020Language Models are Few-Shot Learnerscited by 32,419
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  4. 2022Training language models to follow instructions with human feedbackcited by 15,971
  5. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  6. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  7. 2020Measuring Massive Multitask Language Understandingcited by 7,963
  8. 2020Scaling Laws for Neural Language Modelscited by 6,586
  9. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,355
  10. 2022PaLM: Scaling Language Modeling with Pathwayscited by 5,326
  11. 2022LAION-5B: An open large-scale dataset for training next generation image-text modelscited by 4,152
  12. 2021On the Opportunities and Risks of Foundation Modelscited by 4,033
38 more
  1. 2021Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervisioncited by 3,864
  2. 2021GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Modelscited by 3,614
  3. 2023Generative Agents: Interactive Simulacra of Human Behaviorcited by 3,568
  4. 2022Training Compute-Optimal Large Language Modelscited by 3,019
  5. 2021TruthfulQA: Measuring How Models Mimic Human Falsehoodscited by 2,975
  6. 2022Emergent Abilities of Large Language Modelscited by 2,512
  7. 2019Fine-Tuning Language Models from Human Preferencescited by 2,150
  8. 2013Fine-Grained Visual Classification of Aircraftcited by 1,726
  9. 2021Multitask Prompted Training Enables Zero-Shot Task Generalizationcited by 1,470
  10. 2022Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?cited by 1,434
  11. 2022Reproducible scaling laws for contrastive language-image learningcited by 1,173
  12. 2022Large Language Models Are Human-Level Prompt Engineerscited by 1,045
  13. 2020Exploiting Cloze Questions for Few Shot Text Classification and Natural Language Inferencecited by 938
  14. 2021An Explanation of In-context Learning as Implicit Bayesian Inferencecited by 938
  15. 2016Semantics derived automatically from language corpora contain human-like biasescited by 901
  16. 2019CTRL: A Conditional Transformer Language Model for Controllable Generationcited by 762
  17. 2022Out of One, Many: Using Language Models to Simulate Human Samplescited by 752
  18. 2021FLAVA: A Foundational Language And Vision Alignment Modelcited by 649
  19. 2021Prompt Programming for Large Language Models: Beyond the Few-Shot Paradigmcited by 606
  20. 2017Zero-Shot Learning -- A Comprehensive Evaluation of the Good, the Bad and the Uglycited by 538
  21. 2022Using Large Language Models to Simulate Multiple Humans and Replicate Human Subject Studiescited by 515
  22. 2022InternVideo: General Video Foundation Models via Generative and Discriminative Learningcited by 447
  23. 2022Using cognitive psychology to understand GPT-3cited by 398
  24. 2021Persistent Anti-Muslim Bias in Large Language Modelscited by 377
  25. 2023Toxicity in ChatGPT: Analyzing Persona-assigned Language Modelscited by 360
  26. 2023Causal Reasoning and Large Language Models: Opening a New Frontier for Causalitycited by 353
  27. 2022Visual Classification via Description from Large Language Modelscited by 336
  28. 2022Language in a Bottle: Language Model Guided Concept Bottlenecks for Interpretable Image Classificationcited by 312
  29. 2021An Empirical Study of GPT-3 for Few-Shot Knowledge-Based VQAcited by 301
  30. 2022VQGAN-CLIP: Open Domain Image Generation and Editing with Natural Language Guidancecited by 285
  31. 2023Exploiting Programmatic Behavior of LLMs: Dual-Use Through Standard Security Attackscited by 278
  32. 2022Can language models learn from explanations in context?cited by 249
  33. 2022Emergent Analogical Reasoning in Large Language Modelscited by 240
  34. 2021Understanding the Capabilities, Limitations, and Societal Impact of Large Language Modelscited by 160
  35. 2022Ask Me Anything: A simple strategy for prompting language modelscited by 155
  36. 2023Fundamental Limitations of Alignment in Large Language Modelscited by 155
  37. 2023How well do Large Language Models perform in Arithmetic tasks?cited by 130
  38. 2019On Extractive and Abstractive Neural Document Summarization with Transformer Language Modelscited by 83

50 most cited of 58 All 58 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 20; 2024: 64; 2025: 36; 2026: 35.

155 papers.

Sort:

By area:

  1. 2023The Rise and Potential of Large Language Model Based Agents: A Surveycited by 1,439
  2. 2023AgentVerse: Facilitating Multi-Agent Collaboration and Exploring Emergent Behaviorscited by 598
  3. 2023Challenges and Applications of Large Language Modelscited by 254
  4. 2023Bias Runs Deep: Implicit Reasoning Biases in Persona-Assigned LLMscited by 199
  5. 2024Large Language Models as Urban Residents: An LLM Agent Framework for Personal Mobility Generationcited by 100
  6. 2024MarkLLM: An Open-Source Toolkit for LLM Watermarkingcited by 95
  7. 2024Aligning to Thousands of Preferences via System Message Generalizationcited by 84
  8. 2025LLM Generated Persona is a Promise with a Catchcited by 81
  9. 2023Mark My Words: Analyzing and Evaluating Language Model Watermarkscited by 71
  10. 2024CogBench: a large language model walks into a psychology labcited by 63
  11. 2024One fish, two fish, but not the whole sea: Alignment reduces language models' conceptual diversitycited by 58
  12. 2024Large Language Models Assume People are More Rational than We Really arecited by 54
38 more
  1. 2023Bias Testing and Mitigation in LLM-based Code Generationcited by 53
  2. 2023On the steerability of large language models toward data-driven personascited by 53
  3. 2023Do LLM Agents Exhibit Social Behavior?cited by 53
  4. 2023Large Language Models as Superpositions of Cultural Perspectivescited by 49
  5. 2024Random Silicon Sampling: Simulating Human Sub-Population Opinion Using a Large Language Model Based on Group-Level Demographic Informationcited by 49
  6. 2024Shall We Team Up: Exploring Spontaneous Cooperation of Competing LLM Agentscited by 48
  7. 2024Self-Alignment of Large Language Models via Monopolylogue-based Social Scene Simulationcited by 46
  8. 2023Are Personalized Stochastic Parrots More Dangerous? Evaluating Persona Biases in Dialogue Systemscited by 37
  9. 2024Large Language Models in Cybersecurity: State-of-the-Artcited by 34
  10. 2023Unveiling Gender Bias in Terms of Profession Across LLMs: Analyzing and Addressing Sociological Implicationscited by 29
  11. 2024PICLe: Eliciting Diverse Behaviors from Large Language Models with Persona In-Context Learningcited by 29
  12. 2024Navigating LLM Ethics: Advancements, Challenges, and Future Directionscited by 28
  13. 2024SysBench: Can Large Language Models Follow System Messages?cited by 28
  14. 2024Exploring Safety-Utility Trade-Offs in Personalized Language Modelscited by 27
  15. 2024AgentCoord: Visually Exploring Coordination Strategy for LLM-based Multi-Agent Collaborationcited by 26
  16. 2023Diversity of Thought Improves Reasoning Abilities of LLMscited by 25
  17. 2023Empowering Many, Biasing a Few: Generalist Credit Scoring through Large Language Modelscited by 24
  18. 2024(Ir)rationality and Cognitive Biases in Large Language Modelscited by 23
  19. 2024AutoAD-Zero: A Training-Free Framework for Zero-Shot Audio Descriptioncited by 23
  20. 2024Generative Monoculture in Large Language Modelscited by 22
  21. 2024Helpful assistant or fruitful facilitator? Investigating how personas affect language model behaviorcited by 21
  22. 2024Recent Trends in Personalized Dialogue Generation: A Review of Datasets, Methodologies, and Evaluationscited by 20
  23. 2025An Empirical Analysis of Uncertainty in Large Language Model Evaluationscited by 20
  24. 2025Principled Personas: Defining and Measuring the Intended Effects of Persona Prompting on Task Performancecited by 18
  25. 2024When LLMs Play the Telephone Game: Cultural Attractors as Conceptual Tools to Evaluate LLMs in Multi-turn Settingscited by 17
  26. 2025Unmasking Conversational Bias in AI Multiagent Systemscited by 17
  27. 2024Scene Graph Generation with Role-Playing Large Language Modelscited by 16
  28. 2024Q-Probe: A Lightweight Approach to Reward Maximization for Language Modelscited by 15
  29. 2024Perceptions of Linguistic Uncertainty by Language Models and Humanscited by 15
  30. 2024Stick to your Role! Stability of Personal Values Expressed in Large Language Modelscited by 14
  31. 2024Ask LLMs Directly, "What shapes your bias?": Measuring Social Bias in Large Language Modelscited by 14
  32. 2024Do LLMs have Consistent Values?cited by 14
  33. 2025Whose Personae? Synthetic Persona Experiments in LLM Research and Pathways to Transparencycited by 14
  34. 2024RSBuilding: Towards General Remote Sensing Image Building Extraction and Change Detection with Foundation Modelcited by 12
  35. 2024Self-Prompt Tuning: Enable Autonomous Role-Playing in LLMscited by 12
  36. 2023"One-Size-Fits-All"? Examining Expectations around What Constitute "Fair" or "Good" NLG System Behaviorscited by 11
  37. 2023Smart Agent-Based Modeling: On the Use of Large Language Models in Computer Simulationscited by 11
  38. 2024Your Large Language Model is Secretly a Fairness Proponent and You Should Prompt it Like Onecited by 11

50 most cited of 155 All 155 (JSON)