# Restoring Heterogeneity in LLM-based Social Simulation: An Audience Segmentation Approach

arXiv: https://arxiv.org/abs/2604.06663 · HTML: https://mlsanity.com/paper/2604.06663/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 26 references that are in the map.
- First posted on arXiv 8 Apr 2026.

## Builds on (most cited first, 8 of 8)

- [Generative Agents: Interactive Simulacra of Human Behavior](https://mlsanity.com/paper/2304.03442/) (2023)
- [Out of One, Many: Using Language Models to Simulate Human Samples](https://mlsanity.com/paper/2209.06899/) (2022)
- [Large language models that replace human participants can harmfully misportray and flatten identity groups](https://mlsanity.com/paper/2402.01908/) (2024)
- [Language Model Fine-Tuning on Scaled Survey Data for Predicting Distributions of Public Opinions](https://mlsanity.com/paper/2502.16761/) (2025)
- [Simulating Social Media Using Large Language Models to Evaluate Alternative News Feed Algorithms](https://mlsanity.com/paper/2310.05984/) (2023)
- [Random Silicon Sampling: Simulating Human Sub-Population Opinion Using a Large Language Model Based on Group-Level Demographic Information](https://mlsanity.com/paper/2402.18144/) (2024)
- [Are Large Language Models (LLMs) Good Social Predictors?](https://mlsanity.com/paper/2402.12620/) (2024)
- [LLM-Based Social Simulations Require a Boundary](https://mlsanity.com/paper/2506.19806/) (2025)

## Built on by (most cited first, 4 of 4)

- [Beyond the Mean: Three-Axis Fidelity for Aligning LLM-Based Survey Simulators from Small Pilot Data](https://mlsanity.com/paper/2606.28963/) (2026)
- [Beyond Averages: Evaluating LLMs on Human Survey Replication at the Distributional Level](https://mlsanity.com/paper/2606.09013/) (2026)
- [Emulate or Estimate? The Divergent Strengths of Base and Post-Trained Language Models for Opinion Simulation](https://mlsanity.com/paper/2608.03044/) (2026)
- [When Cosine Similarity Fails to Reflect Linearly Accessible Structure in Dialogue Models](https://mlsanity.com/paper/2609.22522/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) are increasingly used to simulate social attitudes and behaviors, offering scalable "silicon samples" that can approximate human data. However, current simulation practice often collapses diversity into an "average persona," masking subgroup variation that is central to social reality. This study introduces audience segmentation as a systematic approach for restoring heterogeneity in LLM-based social simulation. Using U.S. climate-opinion survey data, we compare six segmentation configurations across two open-weight LLMs (Llama 3.1-70B and Mixtral 8x22B), varying segmentation identifier granularity, parsimony, and selection logic (theory-driven, data-driven, and instrument-based). We evaluate simulation performance with a three-dimensional evaluation framework covering distributional, structural, and predictive fidelity. Results show that increasing identifier granularity does not produce consistent improvement: moderate enrichment can improve performance, but further expansion does not reliably help and can worsen structural and predictive fidelity. Across parsimony comparisons, compact configurations often match or outperform more comprehensive alternatives, especially in structural and predictive fidelity, while distributional fidelity remains metric dependent. Identifier selection logic determines which fidelity dimension benefits most: instrument-based selection best preserves distributional shape, whereas data-driven selection best recovers between-group structure and identifier-outcome associations. Overall, no single configuration dominates all dimensions, and performance gains in one dimension can coincide with losses in another. These findings position audience segmentation as a core methodological approach for valid LLM-based social simulation and highlight the need for heterogeneity-aware evaluation and variance-preserving modeling strategies.