# The Chameleon's Limit: Investigating Persona Collapse and Homogenization in Large Language Models

arXiv: https://arxiv.org/abs/2604.24698 · HTML: https://mlsanity.com/paper/2604.24698/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 33 of its 49 references that are in the map.
- First posted on arXiv 27 Apr 2026.

## Builds on (most cited first, 30 of 33)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [Generative Agents: Interactive Simulacra of Human Behavior](https://mlsanity.com/paper/2304.03442/) (2023)
- [Improving Factuality and Reasoning in Language Models through Multiagent Debate](https://mlsanity.com/paper/2305.14325/) (2023)
- [Understanding Contrastive Representation Learning through Alignment and Uniformity on the Hypersphere](https://mlsanity.com/paper/2005.10242/) (2020)
- [Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debate](https://mlsanity.com/paper/2305.19118/) (2023)
- [Improved Precision and Recall Metric for Assessing Generative Models](https://mlsanity.com/paper/1904.06991/) (2019)
- [Reliable Fidelity and Diversity Metrics for Generative Models](https://mlsanity.com/paper/2002.09797/) (2020)
- [Character-LLM: A Trainable Agent for Role-Playing](https://mlsanity.com/paper/2310.10158/) (2023)
- [Role-Play with Large Language Models](https://mlsanity.com/paper/2305.16367/) (2023)
- [PersonaLLM: Investigating the Ability of Large Language Models to Express Personality Traits](https://mlsanity.com/paper/2305.02547/) (2023)
- [Bias Runs Deep: Implicit Reasoning Biases in Persona-Assigned LLMs](https://mlsanity.com/paper/2311.04892/) (2023)
- [InCharacter: Evaluating Personality Fidelity in Role-Playing Agents through Psychological Interviews](https://mlsanity.com/paper/2310.17976/) (2023)
- [Investigating Cultural Alignment of Large Language Models](https://mlsanity.com/paper/2402.13231/) (2024)
- [Intrinsic Dimension Estimation for Robust Detection of AI-Generated Texts](https://mlsanity.com/paper/2306.04723/) (2023)
- [On the Resilience of LLM-Based Multi-Agent Collaboration with Faulty Agents](https://mlsanity.com/paper/2408.00989/) (2024)
- [Roleplay-doh: Enabling Domain-Experts to Create LLM-simulated Patients via Eliciting and Adhering to Principles](https://mlsanity.com/paper/2407.00870/) (2024)
- [When "A Helpful Assistant" Is Not Really Helpful: Personas in System Prompts Do Not Improve Performances of Large Language Models](https://mlsanity.com/paper/2311.10054/) (2023)
- [The Prompt Makes the Person(a): A Systematic Evaluation of Sociodemographic Persona Prompting for Large Language Models](https://mlsanity.com/paper/2507.16076/) (2025)
- [Consistently Simulating Human Personas with Multi-Turn Reinforcement Learning](https://mlsanity.com/paper/2511.00222/) (2025)
- [LLMs generate structurally realistic social networks but overestimate political homophily](https://mlsanity.com/paper/2408.16629/) (2024)
- [Actions Speak Louder than Words: Agent Decisions Reveal Implicit Biases in Language Models](https://mlsanity.com/paper/2501.17420/) (2025)
- [CoSER: A Comprehensive Literary Dataset and Framework for Training and Evaluating LLM Role-Playing and Persona Simulation](https://mlsanity.com/paper/2502.09082/) (2025)
- [LLM Probability Concentration: How Alignment Shrinks the Generative Horizon](https://mlsanity.com/paper/2506.17871/) (2025)
- [Synthetic Socratic Debates: Examining Persona Effects on Moral Decision and Persuasion Dynamics](https://mlsanity.com/paper/2506.12657/) (2025)
- [VisBias: Measuring Explicit and Implicit Social Biases in Vision Language Models](https://mlsanity.com/paper/2503.07575/) (2025)
- [The PIMMUR Principles: Ensuring Validity in Collective Behavior of LLM Societies](https://mlsanity.com/paper/2509.18052/) (2025)
- [Humanizing Machines: Rethinking LLM Anthropomorphism Through a Multi-Level Framework of Design](https://mlsanity.com/paper/2508.17573/) (2025)

## Built on by (most cited first, 8 of 8)

- [One Frozen Simulator Is Not Enough: Simulator Collapse in Multi-Agent RL](https://mlsanity.com/paper/2608.12253/) (2026)
- [BOUNDARY_SYNC: Measuring Communication-Induced Representational Coupling in Multi-Agent LLM Systems](https://mlsanity.com/paper/2607.01600/) (2026)
- [Co-design of LLM-based preference agents: participation may drive overtrust](https://mlsanity.com/paper/2607.21757/) (2026)
- [When Cosine Similarity Fails to Reflect Linearly Accessible Structure in Dialogue Models](https://mlsanity.com/paper/2609.22522/) (2026)
- [How Far Do Persona Effects Generalize in Language Models?](https://mlsanity.com/paper/2609.32758/) (2026)
- [Persona Following Is Not Selective Control: The Neutrality Gap in LLM User Simulation](https://mlsanity.com/paper/2609.35036/) (2026)
- [Behavioral History Outperforms Descriptions of the Person for LLM Synthetic Personas](https://mlsanity.com/paper/2610.03998/) (2026)
- [Back in Style: A Sociolinguistic Approach to Authoring and Measuring Persona Fidelity in User Simulation](https://mlsanity.com/paper/2610.10988/) (2026)

## Abstract (arXiv)

Applications based on large language models (LLMs), such as multi-agent simulations, require population diversity among agents. We identify a pervasive failure mode we term Persona Collapse: agents each assigned a distinct profile nonetheless converge into a narrow behavioral mode, producing a homogeneous simulated population. To quantify persona collapse, we propose a framework that measures how much of the persona space a population occupies (Coverage), how evenly agents spread across it (Uniformity), and how rich the resulting behavioral patterns are (Complexity). Evaluating ten LLMs on personality simulation (BFI-44), moral reasoning, and self-introduction, we observe persona collapse along two axes: (1) Dimensions: a model can appear diverse on one axis yet structurally degenerate on another, and (2) Domains: the same model may collapse the most in personality yet be the most diverse in moral reasoning. Furthermore, item-level diagnostics reveal that behavioral variation tracks coarse demographic stereotypes rather than the fine-grained individual differences specified in each persona. Counter-intuitively, the models achieving the highest per-persona fidelity consistently produce the most stereotyped populations. We release our toolkit and data to support population-level evaluation of LLMs.