mlsanityData to 7 Oct 2026

AlignSurvey: A Comprehensive Benchmark for Human Preferences Alignment in Social Surveys

arXiv 2511.07871 ↗First posted 11 Nov 2025cs.CLcited by 2

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

Abstract (arXiv)

Understanding human attitudes, preferences, and behaviors through social surveys is essential for academic research and policymaking. Yet traditional surveys face persistent challenges, including fixed-question formats, high costs, limited adaptability, and difficulties ensuring cross-cultural equivalence. While recent studies explore large language models (LLMs) to simulate survey responses, most are limited to structured questions, overlook the entire survey process, and risks under-representing marginalized groups due to training data biases. We introduce AlignSurvey, the first benchmark that systematically replicates and evaluates the full social survey pipeline using LLMs. It defines four tasks aligned with key survey stages: social role modeling, semi-structured interview modeling, attitude stance modeling and survey response modeling. It also provides task-specific evaluation metrics to assess alignment fidelity, consistency, and fairness at both individual and group levels, with a focus on demographic diversity. To support AlignSurvey, we construct a multi-tiered dataset architecture: (i) the Social Foundation Corpus, a cross-national resource with 44K+ interview dialogues and 400K+ structured survey records; and (ii) a suite of Entire-Pipeline Survey Datasets, including the expert-annotated AlignSurvey-Expert (ASE) and two nationally representative surveys for cross-cultural evaluation. We release the SurveyLM family, obtained through two-stage fine-tuning of open-source LLMs, and offer reference models for evaluating domain-specific alignment. All datasets, models, and tools are available at github and huggingface to support transparent and socially responsible research.

Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Placed in Human values and AI feedback alignment, one of 10 directions in LLM alignment / RLHF / preference optimisation. Builds on 18 of its 40 references that are in the map. First posted on arXiv 11 Nov 2025.

Builds on

18 of 40 references.

Closest ancestors · 3

Same direction, newest first

  1. 2025Fair-PP: A Synthetic Dataset for Aligning LLM with Personalized Preferences of Social Equity17 May
  2. 2024COMMUNITY-CROSS-INSTRUCT: Unsupervised Instruction Generation for Aligning Large Language Models to Online Communities17 Jun
  3. 2023Aligning Language Models to User Opinions24 May

Other references · 15

  1. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,767
  2. 2023BeaverTails: Towards Improved Safety Alignment of LLM via a Human-Preference Datasetcited by 828
  3. 2023OpenAssistant Conversations -- Democratizing Large Language Model Alignmentcited by 712
  4. 2024Large language models that replace human participants can harmfully misportray and flatten identity groupscited by 137
  5. 2023Generative Language Models Exhibit Social Identity Biasescited by 107
  6. 2022Moral Mimicry: Large Language Models Produce Moral Rationalizations Tailored to Political Identitycited by 71
  7. 2025Language Model Fine-Tuning on Scaled Survey Data for Predicting Distributions of Public Opinionscited by 64
  8. 2025SocioVerse: A World Model for Social Simulation Powered by LLM Agents and A Pool of 10 Million Real-World Userscited by 50
  9. 2024Virtual Personas for Language Models via an Anthology of Backstoriescited by 49
  10. 2024Persona-DB: Efficient Large Language Model Personalization for Response Prediction with Collaborative Data Refinementcited by 47
  11. 2024Centaur: a foundation model of human cognitioncited by 34
  12. 2024Human and LLM Biases in Hate Speech Annotations: A Socio-Demographic Analysis of Annotators and Targetscited by 31
3 more
  1. 2024Few-shot Personalization of LLMs with Mis-aligned Responsescited by 30
  2. 2024Political-LLM: Large Language Models in Political Sciencecited by 27
  3. 2024PhDGPT: Introducing a psychometric and linguistic dataset about how large language models perceive graduate students and professors in psychologycited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026Assessing the Reliability of Persona-Conditioned LLMs as Synthetic Survey Respondentscited by 7
  2. 2026People Are Not Just Their Countries. Disentangling Social Determinants of LLM Value Alignment Across Europecited by 1

Similar, not cited

Close in wording, no citation either way

  1. 2026Beyond Marginal Distributions: A Framework to Evaluate the Representativeness of Demographic-Aligned LLMs
  2. 2026SurveyReview: A Reviewer-Aligned Benchmark for Survey Evaluators
  3. 2025Emulating Public Opinion: A Proof-of-Concept of AI-Generated Synthetic Survey Responses for the Chilean Case
  4. 2026Measurement Validity in LLM Cultural Alignment
  5. 2025Improving the Distributional Alignment of LLMs using Supervision
  6. 2024The PRISM Alignment Dataset: What Participatory, Representative and Individualised Human Feedback Reveals About the Subjective and Multicultural Alignment of Large Language Models
  7. 2026On the Credibility of Evaluating LLMs using Survey Questions
  8. 2026PALMs: Using Multi Construct-Grounded Rationales for Modeling Population Preferences in LLMs