mlsanityData to 13 Oct 2026

Towards Measuring the Representation of Subjective Global Opinions in Language Models

arXiv 2306.16388 ↗First posted 28 Jun 2023cs.CL cs.AIcited by 428

Not on the map. Why?

Cited by 84 in LLM alignment / RLHF / preference optimisation, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Large language models (LLMs) may not equitably represent diverse global perspectives on societal issues. In this paper, we develop a quantitative framework to evaluate whose opinions model-generated responses are more similar to. We first build a dataset, GlobalOpinionQA, comprised of questions and answers from cross-national surveys designed to capture diverse opinions on global issues across different countries. Next, we define a metric that quantifies the similarity between LLM-generated survey responses and human responses, conditioned on country. With our framework, we run three experiments on an LLM trained to be helpful, honest, and harmless with Constitutional AI. By default, LLM responses tend to be more similar to the opinions of certain populations, such as those from the USA, and some European and South American countries, highlighting the potential for biases. When we prompt the model to consider a particular country's perspective, responses shift to be more similar to the opinions of the prompted populations, but can reflect harmful cultural stereotypes. When we translate GlobalOpinionQA questions to a target language, the model's responses do not necessarily become the most similar to the opinions of speakers of those languages. We release our dataset for others to use and build on. Our data is at https://huggingface.co/datasets/Anthropic/llm_global_opinions. We also provide an interactive visualization at https://llmglobalvalues.anthropic.com.

Key facts. Cited by 428 papers in the map (data through 13 Oct 2026). Builds on 64 of its 95 references that are in the map. First posted on arXiv 28 Jun 2023.

Builds on

64 of 95 references.

References · 64

  1. 2020Language Models are Few-Shot Learnerscited by 32,320
  2. 2021On the Opportunities and Risks of Foundation Modelscited by 3,999
  3. 2017Deep reinforcement learning from human preferencescited by 3,871
  4. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,691
  5. 2022Constitutional AI: Harmlessness from AI Feedbackcited by 3,011
  6. 2020Learning to summarize from human feedbackcited by 2,575
  7. 2019Fine-Tuning Language Models from Human Preferencescited by 2,139
  8. 2022Language Models (Mostly) Know What They Knowcited by 1,823
  9. 2022No Language Left Behind: Scaling Human-Centered Machine Translationcited by 1,302
  10. 2020RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Modelscited by 1,249
  11. 2021Scaling Language Models: Methods, Analysis & Insights from Training Gophercited by 1,201
  12. 2021Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order Sensitivitycited by 1,126
38 more
  1. 2021A General Language Assistant as a Laboratory for Alignmentcited by 1,041
  2. 2022Red Teaming Language Models with Language Modelscited by 1,014
  3. 2021Ethical and social risks of harm from Language Modelscited by 964
  4. 2020Language (Technology) is Power: A Critical Survey of "Bias" in NLPcited by 951
  5. 2020StereoSet: Measuring stereotypical bias in pretrained language modelscited by 918
  6. 2022Discovering Language Model Behaviors with Model-Written Evaluationscited by 824
  7. 2020The State and Fate of Linguistic Diversity and Inclusion in the NLP Worldcited by 792
  8. 2023Whose Opinions Do Language Models Reflect?cited by 768
  9. 2022Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learnedcited by 723
  10. 2020Aligning AI With Shared Human Valuescited by 717
  11. 2019The Woman Worked as a Babysitter: On Biases in Language Generationcited by 540
  12. 2022Improving alignment of dialogue agents via targeted human judgementscited by 529
  13. 2023Benchmarking Large Language Models for News Summarizationcited by 444
  14. 2020Artificial Intelligence, Values and Alignmentcited by 443
  15. 2022ToxiGen: A Large-Scale Machine-Generated Dataset for Adversarial and Implicit Hate Speech Detectioncited by 421
  16. 2019Social Bias Frames: Reasoning about Social and Power Implications of Languagecited by 378
  17. 2021Persistent Anti-Muslim Bias in Large Language Modelscited by 377
  18. 2021AI and the Everything in the Whole Wide World Benchmarkcited by 344
  19. 2021Dealing with Disagreements: Looking Beyond the Majority Vote in Subjective Annotationscited by 308
  20. 2022Social Simulacra: Creating Populated Prototypes for Social Computing Systemscited by 301
  21. 2021Towards Understanding and Mitigating Social Biases in Language Modelscited by 296
  22. 2021Annotators with Attitudes: How Annotator Beliefs And Identities Bias Toxic Language Detectioncited by 261
  23. 2020Data and its (dis)contents: A survey of dataset development and use in machine learning researchcited by 259
  24. 2019Identifying and Reducing Gender Bias in Word-Level Language Modelscited by 214
  25. 2022Predictability and Surprise in Large Generative Modelscited by 209
  26. 2023Marked Personas: Using Natural Language Prompts to Measure Stereotypes in Language Modelscited by 205
  27. 2021Societal Biases in Language Generation: Progress and Challengescited by 202
  28. 2023The political ideology of conversational AI: Converging evidence on ChatGPT's pro-environmental, left-libertarian orientationcited by 188
  29. 2020Social Biases in NLP Models as Barriers for Persons with Disabilitiescited by 182
  30. 2021Process for Adapting Language Models to Society (PALMS) with Values-Targeted Datasetscited by 181
  31. 2023The Capacity for Moral Self-Correction in Large Language Modelscited by 173
  32. 2021Understanding the Capabilities, Limitations, and Societal Impact of Large Language Modelscited by 159
  33. 2023Having Beer after Prayer? Measuring Cultural Bias in Large Language Modelscited by 153
  34. 2023Co-Writing with Opinionated Language Models Affects Users' Viewscited by 148
  35. 2022Probing Pre-Trained Language Models for Cross-Cultural Differences in Valuescited by 147
  36. 2022Jury Learning: Integrating Dissenting Voices into Machine Learning Modelscited by 118
  37. 2022The Ghost in the Machine has an American accent: value conflict in GPT-3cited by 105
  38. 2021Re-imagining Algorithmic Fairness in India and Beyondcited by 104

50 most cited of 64 All 64 (JSON)

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2023: 24; 2024: 128; 2025: 131; 2026: 145.

428 papers.

Sort:

By area:

  1. 2024RewardBench: Evaluating Reward Models for Language Modelingcited by 488
  2. 2023Does Writing with Language Models Reduce Content Diversity?cited by 191
  3. 2023Gender bias and stereotypes in Large Language Modelscited by 187
  4. 2024A Roadmap to Pluralistic Alignmentcited by 179
  5. 2023Do LLMs exhibit human-like response biases? A case study in survey designcited by 177
  6. 2024BLEnD: A Benchmark for LLMs on Everyday Knowledge in Diverse Cultures and Languagescited by 171
  7. 2024Towards Measuring and Modeling "Culture" in LLMs: A Surveycited by 167
  8. 2024Quantifying the Persona Effect in LLM Simulationscited by 166
  9. 2024Political Compass or Spinning Arrow? Towards More Meaningful Evaluations for Values and Opinions in Large Language Modelscited by 162
  10. 2024The PRISM Alignment Dataset: What Participatory, Representative and Individualised Human Feedback Reveals About the Subjective and Multicultural Alignment of Large Language Modelscited by 143
  11. 2024Investigating Cultural Alignment of Large Language Modelscited by 141
  12. 2024Aya Expanse: Combining Research Breakthroughs for a New Multilingual Frontiercited by 128
38 more
  1. 2024Aya 23: Open Weight Releases to Further Multilingual Progresscited by 123
  2. 2024Collective Constitutional AI: Aligning a Language Model with Public Inputcited by 123
  3. 2025Large Language Models Often Know When They Are Being Evaluatedcited by 117
  4. 2024Modular Pluralism: Pluralistic Alignment via Multi-LLM Collaborationcited by 112
  5. 2023Cultural Alignment in Large Language Models: An Explanatory Analysis Based on Hofstede's Cultural Dimensionscited by 111
  6. 2024Survey of Cultural Awareness in Language Models: Text and Beyondcited by 102
  7. 2024CultureBank: An Online Community-Driven Knowledge Base Towards Culturally Aware Language Technologiescited by 98
  8. 2025LLM Social Simulations Are a Promising Research Methodcited by 98
  9. 2023Value Kaleidoscope: Engaging AI with Pluralistic Human Values, Rights, and Dutiescited by 97
  10. 2024Group Robust Preference Optimization in Reward-free RLHFcited by 95
  11. 2024RLHF Deciphered: A Critical Analysis of Reinforcement Learning from Human Feedback for LLMscited by 90
  12. 2024The BiGGen Bench: A Principled Benchmark for Fine-grained Evaluation of Language Models with Language Modelscited by 90
  13. 2024Measuring Political Bias in Large Language Models: What Is Said and How It Is Saidcited by 89
  14. 2024Aligning to Thousands of Preferences via System Message Generalizationcited by 84
  15. 2024Evaluating Large Language Model Biases in Persona-Steered Generationcited by 78
  16. 2024SafetyPrompts: a Systematic Review of Open Datasets for Evaluating and Improving Large Language Model Safetycited by 76
  17. 2024PERSONA: A Reproducible Testbed for Pluralistic Alignmentcited by 71
  18. 2024"My Answer is C": First-Token Probabilities Do Not Match Text Answers in Instruction-Tuned Language Modelscited by 66
  19. 2024Unintended Impacts of LLM Alignment on Global Representationcited by 66
  20. 2025Language Model Fine-Tuning on Scaled Survey Data for Predicting Distributions of Public Opinionscited by 65
  21. 2025NoveltyBench: Evaluating Language Models for Humanlike Diversitycited by 64
  22. 2023Group Preference Optimization: Few-Shot Alignment of Large Language Modelscited by 63
  23. 2024(A)I Am Not a Lawyer, But...: Engaging Legal Experts towards Responsible LLM Policies for Legal Advicecited by 63
  24. 2024WorldValuesBench: A Large-Scale Benchmark Dataset for Multi-Cultural Value Awareness of Language Modelscited by 59
  25. 2024Are Large Language Models Consistent over Value-laden Questions?cited by 58
  26. 2024DailyDilemmas: Revealing Value Preferences of LLMs with Quandaries of Daily Lifecited by 58
  27. 2024DeAL: Decoding-time Alignment for Large Language Modelscited by 55
  28. 2024Beyond prompt brittleness: Evaluating the reliability and consistency of political worldviews in LLMscited by 55
  29. 2024CULTURE-GEN: Revealing Global Cultural Perception in Language Models through Natural Language Promptingcited by 55
  30. 2023On the steerability of large language models toward data-driven personascited by 53
  31. 2025Specializing Large Language Models to Simulate Survey Response Distributions for Global Populationscited by 52
  32. 2025SimBench: Benchmarking the Ability of Large Language Models to Simulate Human Behaviorscited by 52
  33. 2024Understanding the Capabilities and Limitations of Large Language Models for Cultural Commonsensecited by 51
  34. 2025Values in the Wild: Discovering and Analyzing Values in Real-World Language Model Interactionscited by 50
  35. 2024Random Silicon Sampling: Simulating Human Sub-Population Opinion Using a Large Language Model Based on Group-Level Demographic Informationcited by 49
  36. 2023Personas as a Way to Model Truthfulness in Language Modelscited by 47
  37. 2024LLM Voting: Human Choices and AI Collective Decision Makingcited by 47
  38. 2024NormAd: A Framework for Measuring the Cultural Adaptability of Large Language Modelscited by 47

50 most cited of 428 All 428 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025An Analysis of Large Language Models for Simulating User Responses in Surveys
  2. 2025Multilingual != Multicultural: Evaluating Gaps Between Multilingual Capabilities and Cultural Alignment in LLMs
  3. 2025Do Political Opinions Transfer Between Western Languages? An Analysis of Unaligned and Aligned Multilingual LLMs
  4. 2026Understanding Cultural Alignment in Multilingual LLMs via Natural Debate Statements ↗
  5. 2025Debiasing International Attitudes: LLM Agents for Simulating US-China Perception Changes
  6. 2025A Cross-Cultural Comparison of LLM-based Public Opinion Simulation: Evaluating Chinese and U.S. Models on Diverse Societies
  7. 2025Emulating Public Opinion: A Proof-of-Concept of AI-Generated Synthetic Survey Responses for the Chilean Case
  8. 2025Surfacing Subtle Stereotypes: A Multilingual, Debate-Oriented Evaluation of Modern LLMs