mlsanityData to 14 Oct 2026

The Failure Happens Before the Drift: The Social Dynamics of Values in LLM Agent Societies

arXiv 2609.05514 ↗First posted 31 Aug 2026cs.AI cs.MAcited by 0

On the map: LLM agents and tool use → Social simulation and multi-agent debate

Abstract (arXiv)

Large Language Model (LLM)-based agents are increasingly used as proxies for human participants in social science research, yet it remains unclear whether they can faithfully simulate diverse and conflicting human value systems. We present a World Values Survey (WVS)-grounded simulation framework where culturally diverse agents with different communication styles engage in longitudinal, value-laden discussions. Across approximately 4,000 conversations involving 1,200 personas, 15 topics, and three models (GPT-4o, Gemini-2.5-Flash, and Gemma-4-E4B), we evaluate value faithfulness, value drift, and conversational realism. We find that more than 50% of personas fail to express their assigned WVS profiles from the outset, while 2-7% drift after repeated conversations. Ablations removing demographic details improve faithfulness for some models but do not change the broader trend: simulated value distributions still systematically deviate from the assigned WVS profiles. Compared to human discussions, simulated dialogues show a different trade-off between stylistic consistency and semantic diversity, often producing content-wise varied but stylistically repetitive exchanges. These findings suggest that current LLM agents can generate plausible conversations, but remain limited proxies for representing and preserving diverse human value profiles over time.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Social simulation and multi-agent debate, one of 16 directions in LLM agents and tool use. Builds on 18 of its 48 references that are in the map. First posted on arXiv 31 Aug 2026.

Builds on

18 of 48 references.

Closest ancestors · 1

Same direction, newest first

  1. 2023Generative Agents: Interactive Simulacra of Human Behavior7 Apr

Other references · 17

  1. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,170
  2. 2023MemGPT: Towards LLMs as Operating Systemscited by 1,389
  3. 2023Whose Opinions Do Language Models Reflect?cited by 773
  4. 2023SOTOPIA: Interactive Evaluation for Social Intelligence in Language Agentscited by 367
  5. 2023Assessing Cross-Cultural Alignment between ChatGPT and Human Societies: An Empirical Studycited by 206
  6. 2023Do LLMs exhibit human-like response biases? A case study in survey designcited by 180
  7. 2016Winning Arguments: Interaction Dynamics and Persuasion Strategies in Good-faith Online Discussionscited by 110
  8. 2013A Computational Approach to Politeness with Application to Social Factorscited by 106
  9. 2022The Ghost in the Machine has an American accent: value conflict in GPT-3cited by 105
  10. 2025Mind the Value-Action Gap: Do LLMs Act in Alignment with Their Values?cited by 51
  11. 2023EtiCor: Corpus for Analyzing LLMs for Etiquettescited by 33
  12. 2020The Role of Pragmatic and Discourse Context in Determining Argument Impactcited by 18
5 more
  1. 2023Opening up Minds with Argumentative Dialoguescited by 8
  2. 2021Proto: A Neural Cocktail for Generating Appealing Conversationscited by 5
  3. 2026Overstating Attitudes, Ignoring Networks: LLM Biases in Simulating Misinformation Susceptibilitycited by 5
  4. 2026ValueFlow: Measuring the Propagation of Value Perturbations in Multi-Agent LLM Systemscited by 4
  5. 2025User Behavior Prediction as a Generic, Robust, Scalable, and Low-Cost Evaluation Strategy for Estimating Generalization in LLMscited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026ExpertIVS: Sociological Expert Driven Individual Value Simulation in Large Language Models
  2. 2025Value-Based Large Language Model Agent Simulation for Mutual Evaluation of Trust and Interpersonal Closeness
  3. 2026Do LLMs Have Values? A Quantitative Analysis and Alignment Framework for Values in Large Language Models
  4. 2026Alignment by Stereotyping: How LLMs Sacrifice Individual Distinctiveness for Cultural Adaptation
  5. 2025From the Fluency Fallacy to the Micro-to-Macro Validity Gap: Opportunities and Pitfalls of LLMs in Social Simulation
  6. 2025Multilingual Agent-Based World Modeling for Social Science
  7. 2025Are LLM Agents Behaviorally Coherent? Latent Profiles for Social Simulation
  8. 2024Sense and Sensitivity: Evaluating the simulation of social dynamics via Large Language Models