arXiv Computation and Language

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

The paper introduces the concept of Persona Collapse, where large language models (LLMs) produce clustered, non-diverse responses in multi-agent simulations. It defines three diagnostic metrics—Coverage, Uniformity, and Complexity—to quantify this phenomenon across eight LLMs on tasks such as personality, moral reasoning, and self-introduction. Findings show that persona collapse is universal but varies by task, cannot be alleviated by sampling more personas, and that higher per-persona fidelity leads to more polarized populations, while persona-targeted fine-tuning further reduces diversity.

arXiv Computation and Language

Sep 23

PERSONAWEAVER: Controllable Diversity Beyond Conventional Archetypes in Procedural Character Generation

PERSONAWEAVER is a new approach to procedural character generation that separates world building from behavioral specification, using manually curated banks of moral positions and conversational reactions to diversify character behavior. By applying this method across ten realistic and fantastical settings and three large language models, the system produces broader moral and interactional response distributions, varied interpersonal language, response length, sentiment, and less archetypal world attribute combinations compared to prior work.

By Maan Qraitem, Kate Saenko, Bryan A. Plummer

llms

arXiv AI

Jun 18

How Well Do Large Language Models Capture Human Personality?

arXiv:2606. 18263v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used to simulate human populations via persona prompting, often under the assumptions that richer persona descriptions improve behavioral fidelity, similarly sized attribute combinations are equally simulatable, and persona definitions generalize across tasks.

By Aanisha Bhattacharyya, Yaman Kumar Singla, Rajiv Ratn Shah, Changyou Chen, Jitendra Ajmera

llms diffusion safety

arXiv AI

Aug 5

Beyond the Hivemind: Escaping LLM Homogeneity via Meta-Persona Anchoring and Sequential Temperature Scaling

arXiv:2608. 02618v1 Announce Type: new Abstract: Recent studies have identified an ``Artificial Hivemind'' effect in Large Language Models (LLMs) causing models to converge on a narrow, homogenized consensus even for open questions.

By Tairan Fu, Javier Conde, Carlos Arriaga, Gonzalo Mart\'inez, Pedro Reviriego, Javier Coronado-Bl\'azquez

llms benchmarks

arXiv AI

Jul 24

More Is Not More: What Matters for Diversity in LLM Opinions?

arXiv:2607. 20429v1 Announce Type: cross Abstract: Large language models are increasingly used to simulate diverse human opinions in open-ended tasks such as synthetic surveys, focus group modeling, and public opinion prediction.

By Qiyang Yao

llms

arXiv Computation and Language

Oct 8

Beyond Cooperative Simulators: Generating Realistic User Personas for Robust Evaluation of LLM Agents

arXiv:2605.12894v2 Announce Type: replace-cross Abstract: Large Language Model (LLM) agents are increasingly deployed in settings where they interact with diverse users, including those who are uncle...

By Harshita Chopra, Kshitish Ghate, Aylin Caliskan, Tadayoshi Kohno, Chirag Shah, Natasha Jaques

llms agents fine-tuning benchmarks

arXiv Computation and Language

Sep 22

Pretrained Persona Mixture Models and Tandem Models for Human Simulation

arXiv:2609.22607v1 Announce Type: new Abstract: We argue here that the current dominant practice in LLM human simulation: prompting instruction-tuned assistant language models to role-play personas,...

By Minwoo Kang, T\'ea Wright, Seun Eisape, Ayush Raj, Suhong Moon, Joseph Suh, Alane Suhr, David M. Chan, John Canny

llms

arXiv AI

Oct 15

Agent Sociodrama: Social-Dynamics-Guided Persona Recasting for LLM Agent Teams

arXiv:2610.15788v1 Announce Type: new Abstract: Large language model (LLM) agents are increasingly deployed as teams, yet they fail in characteristically social ways: information holders stay silent,...

By Yirui Qi, Xiaoming Zhang, Jaehong Yoon

llms agents benchmarks

arXiv Computation and Language

Aug 21

Mitigating Identity Essentialism in LLM Agents with Longitudinal Life Trajectories

arXiv:2608. 19621v1 Announce Type: new Abstract: Large language models (LLMs) offer a scalable approach to social simulation, but their credibility depends on how agents are constructed.

By Hexi Wang, Yujia Zhou, Bangde Du, Weihang Su, Xinyuan Cao, Qingyi Pan, Qingyao Ai, Yueyue Wu, Min Zhang, Yiqun Liu

llms agents safety

arXiv AI

Oct 15

Behavioral History Outperforms Descriptions of the Person for LLM Synthetic Personas

arXiv:2610.03998v2 Announce Type: replace Abstract: Large language models (LLMs) are increasingly used as synthetic personas representing survey respondents. Their validity as substitutes for particu...

By Khashayar Pourtaheri, Ahmad Zareei

llms benchmarks safety

arXiv AI

Jun 17

Patients With Personality: Realistic Patient Simulation through Controlled Diversity and Selective Disclosure

arXiv:2606. 17441v1 Announce Type: cross Abstract: Simulating realistic patient interactions is a key requirement to testing clinical applications of LLMs at scale without time-consuming and expensive user studies.

By Moritz Schlager, Friederike Jungmann, Samuel Schmidgall, Philipp Raffler, Franziska Hartl, Eva Wende, Paula Ro{\ss}m\"uller, Conrad Ketzer, Avinatan Hassidim, Dale R. Webster, Yossi Matias, Yun Liu, Daniel Rueckert, Mike Schaekermann, Paul Hager

llms benchmarks

arXiv AI

Sep 30

Population Fidelity: Evaluating Population Representativeness in LLMs

The paper introduces Population Fidelity, an evaluation framework for assessing how well large language models (LLMs) represent human population attitudes. It focuses on three dimensions: group-level accuracy, between-group variation, and the structure of that variation. Using the framework, the authors replicate a prior study on machine bias and test cultural fine-tuning, finding that while fine-tuning improves overall alignment, it does not enhance representation of within-population differences.

By Neemias B. da Silva, Martin Lukk, Ali Sutani, Abhishek Moturu, Harris Yang, Daniel Silver, Matt Ratto, Thiago H. Silva

llms fine-tuning safety

arXiv AI

Sep 10

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

The study introduces a World Values Survey–grounded simulation framework to test whether large language model agents can faithfully represent diverse human value systems. In about 4,000 conversations with 1,200 personas across three models, more than half of the agents failed to express their assigned value profiles from the start, and only 2–7% drifted over time. The results show systematic deviations from the intended value distributions and reveal that simulated dialogues differ from human discussions in their balance of stylistic consistency and semantic diversity.

By Farah Atif, Sougata Saha, Monojit Choudhury

llms agents

Related stories