paper-with-me

Papers

Synonymix: Unified Group Personas for Generative Simulations

2026-03-30 · Huanxing Chen, Aditesh Kumar arxiv

Generative agent simulations operate at two scales: individual personas for character interaction, and population models for collective behavior analysis and intervention testing. We propose a third scale: meso-level simulation - interaction with group-level representations that retain grounding in rich individual experience. To enable this, we present Synonymix, a pipeline that constructs a "unigraph" from multiple life story personas via graph-based abstraction and merging, producing a queryable collective representation that can be explored for sensemaking or sampled for synthetic persona generation. Evaluating synthetic agents on General Social Survey items, we demonstrate behavioral signal preservation beyond demographic baselines (p<0.001, r=0.59) with demonstrable privacy guarantee (max source contribution <13%). We invite discussion on interaction modalities enabled by meso-level simulations, and whether "high-fidelity" personas can ever capture the texture of lived experience.

📄 PDF Abstract BibTeX arXiv:2603.28066

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Implicit Behavioral Alignment of Language Agents in High-Stakes Crowd Simulations

2025-09-19 · Yunzhe Wang, Gale M. Lucas, Burcin Becerik-Gerber, Volkan Ustun arxiv

Language-driven generative agents have enabled large-scale social simulations with transformative uses, from interpersonal training to aiding global policy-making. However, recent studies indicate that generative agent b…

HACHIMI: Scalable and Controllable Student Persona Generation via Orchestrated Agents

2026-03-05 · Yilin Jiang, Fei Tan, Xuanyu Yin, Jing Leng 외 arxiv

Student Personas (SPs) are emerging as infrastructure for educational LLMs, yet prior work often relies on ad-hoc prompting or hand-crafted profiles with limited control over educational theory and population distributio…

Marked Personas: Using Natural Language Prompts to Measure Stereotypes in Language Models

2023-05-29 · Myra Cheng, Esin Durmus, Dan Jurafsky

To recognize and mitigate harms from large language models (LLMs), we need to understand the prevalence and nuances of stereotypes in LLM outputs. Toward this end, we present Marked Personas, a prompt-based method to mea…

Story Generation

Consistently Simulating Human Personas with Multi-Turn Reinforcement Learning

2025-10-31 · Marwa Abdulhai, Ryan Cheng, Donovan Clay, Tim Althoff 외 arxiv

Large Language Models (LLMs) are increasingly used to simulate human users in interactive settings such as therapy, education, and social role-play. While these simulations enable scalable training and evaluation of AI a…

Reinforcement Learning

German General Social Survey Personas: A Survey-Derived Persona Prompt Collection for Population-Aligned LLM Studies

2025-11-19 · Jens Rupprecht, Leon Fröhling, Claudia Wagner, Markus Strohmaier arxiv

The use of Large Language Models (LLMs) for simulating human perspectives via persona prompting is gaining traction in computational social science. However, well-curated, empirically grounded persona collections remain …