Elamur Research⌕
☰
home / papers / 2026 / ExpertIVS: Sociological Expert Driven Individual Value Simulation in Large Language Models

ExpertIVS: Sociological Expert Driven Individual Value Simulation in Large Language Models

Paper · 2026 · Zhen Wang, Yuqi Ren, Yuehan Cui et al.

Summary

LLM agents used for social simulation often fail to reproduce how a specific person thinks about values. Typical approaches paste survey answers into prompts as isolated statements, which loses the links between value dimensions. Evaluation usually relies on re-administering multiple-choice questionnaires, which ignores how values appear in live conversation. ExpertIVS addresses both problems. It builds 14 LLM-based "sociological expert" agents, one per World Values Survey (WVS) block. These experts read a respondent's answers and turn them into coherent narrative profiles. Each profile then becomes the system prompt for a simulated individual. To test these individuals in interaction, the authors design a roundtable debate protocol with three metrics. In experiments with 480 real respondents from 12 countries, ExpertIVS reproduced original WVS answers with high fidelity. It also outperformed the IndieValue-style baseline when predicting held-out items. In debates, its agents kept their positions and showed differences in attitude tied to their backgrounds. This makes ExpertIVS a more realistic, value-grounded basis for social simulation research.

Key points

  • The 14 expert agents cover 13 WVS value dimensions plus Demographics. Examples include Social Capital & Trust, Migration, Religious Values, Postmaterialist Index, and Political Culture.
  • Questionnaire refilling reached 90.78% value restoration fidelity. Leave-One-Out generalization accuracy improved by 5.3% over the baseline.
  • In the debate setup, 8 simulated individuals, a base LLM (e.g., Gemini 2.5-flash), and a moderator discuss a topic over 40 rounds. Three measures are used to evaluate performance:
    • VAS (Value Alignment Score)
    • SSS (Style Simulation Score)
    • PDI (Persona Distinctiveness Index)
  • Compared with IndieValue, SSS rose from 32.03 to 50.14 on "Accept Newcomers" and from 24.13 to 47.25 on "Local or Universal". Human judgments correlated with the LLM scores (Spearman ρ = 0.826 for VAS, 0.782 for SSS).
  • Ablations show that demographic attributes add little to value alignment. A remaining limitation is that agents still struggle to reproduce authentic sociolects and default to standard written style.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓Tianjin UniversityTianjin · China41 works · 1,128 citations🎓Tianjin Normal UniversityTianjin · China2 works · 11 citations

 Fields, topics and concepts

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 works

🧭 Topics 1Research subjects and languages the work is about.Language models & pretraining · 3,320 works

Cite

Cite as Zhen Wang, Yuqi Ren, Yuehan Cui, Hongxiang Wang, Jianxiang Peng, Zhaoxia Zhang, Bingkun Zhu, Tongxuan Zhang, Dezhi Tong and Deyi Xiong. 2026. ExpertIVS: Sociological Expert Driven Individual Value Simulation in Large Language Models. Findings of the Association for Computational Linguistics: ACL 2026.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.findings-acl.932

@inproceedings{wang2026expertivs,
  title = {{ExpertIVS: Sociological Expert Driven Individual Value Simulation in Large Language Models}},
  author = {Zhen Wang, Yuqi Ren, Yuehan Cui, Hongxiang Wang, Jianxiang Peng, Zhaoxia Zhang, Bingkun Zhu, Tongxuan Zhang, Dezhi Tong and Deyi Xiong},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.932},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.932}
}

References

36 references, 8 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Yuqi Bai, Tianyu Huang, Kun Sun, and Yuting Chen. 2025. Scaling law in llm simulated personality: More detailed and realistic persona profile is all you need. arXiv preprint arXiv:2510.11734. — [arXiv:2510.11734](https://arxiv.org/abs/2510.11734) find
  2. Louis Castricato, Nathan Lile, Rafael Rafailov, Jan-Philipp Fränken, and Chelsea Finn. 2025. Persona: A reproducible testbed for pluralistic alignment. In Proceedings of the 31st International Conference on Computational Linguistics, pages 11348–11368. find
  3. Yu Ying Chiu, Liwei Jiang, Bill Yuchen Lin, Chan Young Park, Shuyue Stella Li, Sahithya Ravi, Mehar Bhatia, Maria Antoniak, Yulia Tsvetkov, Vered Shwartz, and 1 others. 2024. Culturalbench: A robust, diverse, and challenging cultural benchmark by human-ai culturalteaming. arXiv preprint arXiv:2410.0… — [arXiv:2410.02677](https://arxiv.org/abs/2410.02677) find
  4. Gordon Dai, Weijia Zhang, Jinhan Li, Siqi Yang, Srihas Rao, Arthur Caetano, Misha Sra, and 1 others. 2024. Artificial leviathan: Exploring social evolution of llm agents through the lens of hobbesian social contract theory. arXiv preprint arXiv:2406.14373. — [arXiv:2406.14373](https://arxiv.org/abs/2406.14373) find
  5. Susan T Fiske and Steven L Neuberg. 1990. A continuum of impression formation, from category-based to individuating processes: Influences of information and motivation on attention and interpretation. In Advances in experimental social psychology, volume 23, pages 1–74. Elsevier. find
  6. Chen Gao, Xiaochong Lan, Zhihong Lu, Jinzhu Mao, Jinghua Piao, Huandong Wang, Depeng Jin, and Yong Li. 2023. S3: Social-network simulation system with large language model-empowered agents. arXiv preprint arXiv:2307.14984. — [arXiv:2307.14984](https://arxiv.org/abs/2307.14984) find
  7. Tao Ge, Xin Chan, Xiaoyang Wang, Dian Yu, Haitao Mi, and Dong Yu. 2024. Scaling synthetic data creation with 1,000,000,000 personas. arXiv preprint arXiv:2406.20094.
  8. Christian Haerpfer, Ronald Inglehart, Alejandro Moreno, Christian Welzel, Kseniya Kizilova, Jaime Diez-Medrano, Marta Lagos, Pippa Norris, Eduard Ponarin, Bjorn Puranen, and 1 others. 2022. World values survey: Round seven–country-pooled datafile version 5.0. find
  9. Seungwook Han, Idan Shenfeld, Akash Srivastava, Yoon Kim, and Pulkit Agrawal. 2024. Value augmented sampling for language model alignment and personalization. arXiv preprint arXiv:2405.06639. — [arXiv:2405.06639](https://arxiv.org/abs/2405.06639) find
  10. Joel Jang, Seungone Kim, Bill Yuchen Lin, Yizhong Wang, Jack Hessel, Luke Zettlemoyer, Hannaneh Hajishirzi, Yejin Choi, and Prithviraj Ammanabrolu. 2023. Personalized soups: Personalized large language model alignment via post-hoc parameter merging. arXiv preprint arXiv:2310.11564. find
  11. Guangyuan Jiang, Manjie Xu, Song-Chun Zhu, Wenjuan Han, Chi Zhang, and Yixin Zhu. 2023. Evaluating and inducing personality in pre-trained language models. Advances in Neural Information Processing Systems, 36:10622–10643. find
  12. Liwei Jiang, Taylor Sorensen, Sydney Levine, and Yejin Choi. 2025. Can language models reason about individualistic human values and preferences? In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 6757–6794.
  13. Dayeon Ki, Rachel Rudinger, Tianyi Zhou, and Marine Carpuat. 2025. Multiple llm agents debate for equitable cultural alignment. arXiv preprint arXiv:2505.24671.
  14. Hannah Rose Kirk, Bertie Vidgen, Paul Röttger, and Scott A Hale. 2024. The benefits, risks and bounds of personalizing the alignment of large language models to individuals. Nature Machine Intelligence, 6(4):383–392. find
  15. Louis Kwok, Michal Bravansky, and Lewis D Griffin. 2024. Evaluating cultural adaptability of a large language model via simulation of synthetic personas. arXiv preprint arXiv:2408.06929. — [arXiv:2408.06929](https://arxiv.org/abs/2408.06929) find
  16. Zhigen Li, Jianxiang Peng, Yanmeng Wang, Yong Cao, Tianhao Shen, Minghui Zhang, Linxi Su, Shang Wu, Yihang Wu, Yuqian Wang, and 1 others. 2025. Chatsop: An sop-guided mcts planning framework for controllable llm dialogue agents. In Proceedings of the 63rd Annual Meeting of the Association for Comput… find
  17. Marlene Lutz, Indira Sen, Georg Ahnert, Elisa Rogers, and Markus Strohmaier. 2025. The prompt makes the person (a): A systematic evaluation of sociodemographic persona prompting for large language models. arXiv preprint arXiv:2507.16076. find
  18. Andrew Miles and Catherine Yeh. 2022. Do demographic predictors of personal values vary by context? a test of schwartz’s value development theory. Social Sciences & Humanities Open, 5(1):100264. find
  19. Suhong Moon, Marwa Abdulhai, Minwoo Kang, Joseph Suh, Widyadewi Soedarmadji, Eran Kohen Behar, and David M Chan. 2024. Virtual personas for language models via an anthology of backstories. arXiv preprint arXiv:2407.06576. find
  20. Xinyi Mou, Xuanwen Ding, Qi He, Liang Wang, Jingcong Liang, Xinnong Zhang, Libo Sun, Jiayu Lin, Jie Zhou, Xuanjing Huang, and 1 others. 2024. From individual to society: A survey on social simulation driven by large language model-based agents. arXiv preprint arXiv:2412.03563. — [arXiv:2412.03563](https://arxiv.org/abs/2412.03563) find
  21. Vera Neplenbroek, Arianna Bisazza, and Raquel Fernández. 2025. Reading between the prompts: How stereotypes shape llm’s implicit personalization. arXiv preprint arXiv:2505.16467.
  22. Joon Sung Park, Joseph O’Brien, Carrie Jun Cai, Meredith Ringel Morris, Percy Liang, and Michael S Bernstein. 2023. Generative agents: Interactive simulacra of human behavior. In Proceedings of the 36th annual acm symposium on user interface software and technology, pages 1–22. find
  23. Joon Sung Park, Carolyn Q Zou, Aaron Shaw, Benjamin Mako Hill, Carrie Cai, Meredith Ringel Morris, Robb Willer, Percy Liang, and Michael S Bernstein. 2024. Generative agent simulations of 1,000 people. arXiv preprint arXiv:2411.10109. find
  24. Jianxiang Peng, Ling Shi, Xinwei Wu, Hanwen Zhang, Fujiang Liu, Haocheng Lyu, and Deyi Xiong. 2025. Diplomacyagent: Do llms balance interests and ethical principles in international events? In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 13732–13750. find
  25. Jinghua Piao, Yuwei Yan, Jun Zhang, Nian Li, Junbo Yan, Xiaochong Lan, Zhihong Lu, Zhiheng Zheng, Jing Yi Wang, Di Zhou, and 1 others. 2025. Agentsociety: Large-scale simulation of llm-driven generative agents advances understanding of human behaviors and society. arXiv preprint arXiv:2502.08691. — [arXiv:2502.08691](https://arxiv.org/abs/2502.08691) find
  26. Shaoyang Xu, Weilong Dong, Zishan Guo, Xinwei Wu, and Deyi Xiong. 2024. Exploring multilingual concepts of human values in large language models: Is value alignment consistent, transferable and controllable across languages? In Findings of the Association for Computational Linguistics: EMNLP 2024, p… find
  27. Shaoyang Xu, Yongqi Leng, Linhao Yu, and Deyi Xiong. 2025. Self-pluralising culture alignment for large language models. In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers),… find
  28. Xuhui Zhou, Hao Zhu, Leena Mathur, Ruohong Zhang, Haofei Yu, Zhengyang Qi, Louis-Philippe Morency, Yonatan Bisk, Daniel Fried, Graham Neubig, and 1 others. 2023. Sotopia: Interactive evaluation for social intelligence in language agents. arXiv preprint arXiv:2310.11667.
  29. Minjun Zhu, Yixuan Weng, Linyi Yang, and Yue Zhang. 2024. Personality alignment of large language models. arXiv preprint arXiv:2408.11779.
  30. Yiting Ran, Xintao Wang, Tian Qiu, Jiaqing Liang, Yanghua Xiao, and Deqing Yang. 2025. Bookworld: From novels to interactive agent societies for story creation. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 15898–15912. find
  31. Pratik S Sachdeva and Tom van Nuenen. 2025. Deliberative dynamics and value alignment in llm debates. arXiv preprint arXiv:2510.10002. — [arXiv:2510.10002](https://arxiv.org/abs/2510.10002) find
  32. Hua Shen, Nicholas Clark, and Tanu Mitra. 2025. Mind the value-action gap: Do llms act in alignment with their values? In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 3097–3118.
  33. Dan Shi, Tianhao Shen, Yufei Huang, Zhigen Li, Yongqi Leng, Renren Jin, Chuang Liu, Xinwei Wu, Zishan Guo, Linhao Yu, and 1 others. 2024. Large language model safety: A holistic survey. arXiv preprint arXiv:2412.17686. — [arXiv:2412.17686](https://arxiv.org/abs/2412.17686) find
  34. Pia Sommerauer, Giulia Rambelli, and Tommaso Caselli. 2025. Simulating identity, propagating bias: Abstraction and stereotypes in llm-generated text. arXiv preprint arXiv:2509.08484. find
  35. Amir Taubenfeld, Yaniv Dover, Roi Reichart, and Ariel Goldstein. 2024. Systematic biases in llm simulations of debates. arXiv preprint arXiv:2402.04049.
  36. Daniel Wang, Eli Brignac, Minjia Mao, and Xiao Fang. 2025. Measuring stereotype and deviation biases in large language models. arXiv preprint arXiv:2508.06649. — [arXiv:2508.06649](https://arxiv.org/abs/2508.06649) find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/wang-2026-expertivs-sociological-expert-driven-individual/