mlsanityData to 14 Oct 2026

Cultivating Pluralism In Algorithmic Monoculture: The Community Alignment Dataset

arXiv 2507.09650 ↗First posted 13 Jul 2025cs.LGcited by 35

On the map: LLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment

Abstract (arXiv)

How can large language models (LLMs) serve users with varying preferences that may conflict across cultural, political, or other dimensions? To advance this challenge, this paper establishes four key results. First, we demonstrate, through a large-scale multilingual human study with representative samples from five countries (N=15,000), that humans exhibit substantially more variation in preferences than the responses of 21 state-of-the-art LLMs. Second, we show that existing methods for preference dataset collection are insufficient for learning the diversity of human preferences even along two of the most salient dimensions of variability in global values, due to the underlying homogeneity of candidate responses. Third, we argue that this motivates the need for negatively-correlated sampling when generating candidate sets, and we show that simple prompt-based techniques for doing so greatly enhance the performance of alignment methods in learning heterogeneous preferences. Fourth, based on this novel candidate sampling approach, we collect and open-source Community Alignment} the largest and most representative multilingual and multi-turn preference dataset to date, featuring 233,319 comparisons from annotators spanning five countries. Overall, we hope that the Community Alignment dataset will be a valuable resource for improving the effectiveness of LLMs for a diverse global population.

Key facts. Cited by 35 papers in the map (data through 14 Oct 2026). Placed in Human values and AI feedback alignment, one of 10 directions in LLM alignment / RLHF / preference optimisation. Builds on 34 of its 100 references that are in the map. First posted on arXiv 13 Jul 2025.

Builds on

34 of 100 references.

Closest ancestors · 6

Same direction, newest first

  1. 2025LoRe: Personalizing LLMs via Low-Rank Reward Modeling20 Apr
  2. 2024Personalized Adaptation via In-Context Preference Learning17 Oct
  3. 2024Language Model Alignment in Multilingual Trolley Problems2 Jul
  4. 2024Personalized Language Modeling from Personalized Human Feedback6 Feb
  5. 2023Personalized Soups: Personalized Large Language Model Alignment via Post-hoc Parameter Merging17 Oct
  6. 2022Constitutional AI: Harmlessness from AI Feedback15 Dec

Other references · 28

  1. 2022Training language models to follow instructions with human feedbackcited by 15,971
  2. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  3. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,720
  4. 2023Towards Measuring the Representation of Subjective Global Opinions in Language Modelscited by 437
  5. 2024PKU-SafeRLHF: Towards Multi-Level Safety Alignment for LLMs with Human Preferencecited by 261
  6. 2024BLEnD: A Benchmark for LLMs on Everyday Knowledge in Diverse Cultures and Languagescited by 173
  7. 2023Increasing Diversity While Maintaining Accuracy: Text Data Generation with Large Language Models and Human Interventionscited by 125
  8. 2025What Makes a Reward Model a Good Teacher? An Optimization Perspectivecited by 73
  9. 2023Particle Guidance: non-I.I.D. Diverse Sampling with Diffusion Modelscited by 71
  10. 2024Massively Multi-Cultural Knowledge Acquisition & LM Benchmarkingcited by 63
  11. 2025Randomness, Not Representation: The Unreliability of Evaluating Cultural Alignment in LLMscited by 54
  12. 2025Diverse Preference Optimizationcited by 52
16 more
  1. 2023Everyone Deserves A Reward: Learning Customized Human Preferencescited by 43
  2. 2025Multilingual != Multicultural: Evaluating Gaps Between Multilingual Capabilities and Cultural Alignment in LLMscited by 39
  3. 2025Is a Good Foundation Necessary for Efficient Reinforcement Learning? The Computational Role of the Base Model in Explorationcited by 36
  4. 2023Assessing LLMs for Moral Value Pluralismcited by 34
  5. 2025The Shrinking Landscape of Linguistic Diversity in the Age of Large Language Modelscited by 29
  6. 2024Aligning Crowd Feedback via Distributional Preference Reward Modelingcited by 28
  7. 2024Generative Monoculture in Large Language Modelscited by 22
  8. 2022Cultural Incongruencies in Artificial Intelligencecited by 21
  9. 2023Democratic Policy Development using Collective Dialogues and AIcited by 17
  10. 2025Jackpot! Alignment as a Maximal Lotterycited by 17
  11. 2024CIVICS: Building a Dataset for Examining Culturally-Informed Values in Large Language Modelscited by 16
  12. 2023'Generative CI' through Collective Response Systemscited by 13
  13. 2025The Value of Disagreement in AI Design, Evaluation, and Alignmentcited by 12
  14. 2025Societal and technological progress as sewing an ever-growing, ever-changing, patchy, and polychrome quiltcited by 9
  15. 2024Data-Centric Human Preference with Rationales for Direct Preference Alignmentcited by 5
  16. 2025Preference learning made easy: Everything should be understood through win ratecited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 6; 2026: 29.

35 papers.

Sort:

By area:

  1. 2025What's In My Human Feedback? Learning Interpretable Descriptions of Preference Datacited by 26
  2. 2025Whose View of Safety? A Deep DIVE Dataset for Pluralistic Alignment of Text-to-Image Modelscited by 11
  3. 2025Value Drifts: Tracing Value Alignment During LLM Post-Trainingcited by 10
  4. 2025Operationalizing Pluralistic Values in Large Language Model Alignment Reveals Trade-offs in Safety, Inclusivity, and Model Behaviorcited by 9
  5. 2026Strategic Algorithmic Monoculture: Experimental Evidence from Coordination Gamescited by 7
  6. 2026PluriHarms: Benchmarking the Full Spectrum of Human Judgments on AI Harmcited by 5
  7. 2026Positive Alignment: Artificial Intelligence for Human Flourishingcited by 5
  8. 2025Identity-Aware Large Language Models require Cultural Reasoningcited by 4
  9. 2025EVALUESTEER: Measuring Reward Model Steerability Towards Values and Preferencescited by 3
  10. 2026PERSPECTRA: A Scalable and Configurable Pluralist Benchmark of Perspectives from Argumentscited by 3
  11. 2026Parametric Social Identity Injection and Diversification in Public Opinion Simulationcited by 2
  12. 2026Topics as Proxies for Sociodemographics: How Conversational Context Affects LLM Answerscited by 2
23 more
  1. 2026Humans and LLMs Diverge on Probabilistic Inferencescited by 1
  2. 2026Automated Concept Discovery for LLM-as-a-Judge Preference Analysiscited by 1
  3. 2026SCRuB: Social Concept Reasoning under Rubric-Based Evaluationcited by 1
  4. 2026PRISM-X: Experiments on Personalised Fine-Tuning with Human and Simulated Userscited by 1
  5. 2026Learning to Route Languages for Multilingual Policy Optimizationcited by 1
  6. 2026Benchmarking Frontier LLMs on Arabic Cultural and Sociolinguistic Knowledge: A Cross-Evaluation Framework with Human SME Ground Truthcited by 1
  7. 2026Mitigating Identity Essentialism in LLM Agents with Longitudinal Life Trajectoriescited by 1
  8. 2026Distortion of AI Alignment Revisited: RLHF is a Decent Utilitarian Alignercited by 1
  9. 2026CuMA: Aligning LLMs with Sparse Cultural Values via Demographic-Aware Mixture of Adapterscited by 0
  10. 2026How Value Induction Reshapes LLM Behaviourcited by 0
  11. 2026Quantifying the Salience of Geo-Cultural Values for Pluralistic Safety Alignmentcited by 0
  12. 2026Pigeonholing: how bad prompts hurt models, causing collapse and mistakescited by 0
  13. 2026PEBS: Per-rater Empirical-Bayes Shrinkage for RLHF Reward-Model Calibrationcited by 0
  14. 2026A Roadmap to Impactful Pluralistic Alignment Researchcited by 0
  15. 2026What do Reward Models Memorize?cited by 0
  16. 2026PALMs: Using Multi Construct-Grounded Rationales for Modeling Population Preferences in LLMscited by 0
  17. 2026Where Models Converge and Humans Diverge: A Coverage Framework for Distributional Pluralism in Open-Ended Generationcited by 0
  18. 2026Group Alignment-Induced Sycophancy: A Two-Sided Evaluation of Steerable Pluralistic Alignmentcited by 0
  19. 2026Algorithmic Impact Reveals the Hidden Social Choice Structure of Alignmentcited by 0
  20. 2026Monocultural Biases: Correlated biases in large language models lead to unequal systemic exclusion rates in hiringcited by 0
  21. 2026Fair Like Us? Auditing LLM Alignment in Resource Allocationcited by 0
  22. 2026Demographic Pluralism: Inference-Time Modeling of Pluralistic Human Preference Distributionscited by 0
  23. 2026RealCompanion: Benchmarking Human Understanding from Reasoning over Longitudinal Real-World Conversationscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Towards a Unified View of Preference Learning for Large Language Models: A Survey
  2. 2023On Diversified Preferences of Large Language Model Alignment
  3. 2024Modular Pluralism: Pluralistic Alignment via Multi-LLM Collaboration
  4. 2026Aligned but Flattened: Analyzing the Trade-off between Cultural Alignment and Diversity in LLMs
  5. 2026Can Persona-Prompted LLMs Emulate Subgroup Values? An Empirical Analysis of Generalisability and Fairness in Cultural Alignment
  6. 2024A Survey on Human Preference Learning for Large Language Models
  7. 2023Group Preference Optimization: Few-Shot Alignment of Large Language Models
  8. 2025Latent Preference Coding: Aligning Large Language Models via Discrete Latent Codes