LLM Simulation

LLM simulation uses large language models to create computational models of human behavior and social phenomena, aiming to understand and potentially mitigate complex societal issues. Current research focuses on simulating opinion dynamics in social networks, exploring the impact of demographic and social factors (personas) on LLM outputs, and analyzing biases inherent in these simulations, often employing techniques like persona prompting and self-fine-tuning. This approach offers valuable insights into human behavior and social processes, with applications ranging from improving machine translation in low-resource languages to enhancing the interpretability of medical diagnoses and potentially informing strategies for addressing social polarization.

Papers

October 30, 2024

October 24, 2024

From a Tiny Slip to a Giant Leap: An LLM-Based Simulation for Fake News Evolution
Yuhan Liu, Zirui Song, Xiaoqing Zhang, Xiuying Chen, Rui Yan
Real World Fake News Specie Evolution Leap Forward Continuous Network LLM Simulation

September 28, 2024

Decoding Echo Chambers: LLM-Powered Simulations Revealing Polarization in Social Networks
Chenxi Wang, Zongfang Liu, Dequan Yang, Xiuying Chen
Social Network Political Polarization Opinion Dynamic Echo Chamber Opinion Polarization Opinion Aggregation LLM Simulation

March 4, 2024

adaptMLLM: Fine-Tuning Multilingual Language Models on Low-Resource Languages with Integrated LLM Playgrounds
Séamus Lankford, Haithem Afli, Andy Way
Low Resource Language Multilingual Language Model MLLM Attention LLM Simulation

February 16, 2024

Quantifying the Persona Effect in LLM Simulations
Tiancheng Hu, Nigel Collier
Large Language Model Human Annotation Persona Consistency LLM Simulation

February 6, 2024

Systematic Biases in LLM Simulations of Debates
Amir Taubenfeld, Yaniv Dover, Roi Reichart, Ariel Goldstein
Social Bias Statistical Learning Debate Evaluation Social Dynamic Systematic Bias LLM Simulation

December 24, 2023

The Challenge of Using LLMs to Simulate Human Behavior: A Causal Inference Perspective
George Gui, Olivier Toubia
Large Language Model Causal Inference Challenge Task Human Behavior Conditional Average Treatment Effect Global Endogenous Variate LLM Simulation

October 17, 2023

CoMPosT: Characterizing and Evaluating Caricature in LLM Simulations
Myra Cheng, Tiziano Piccardi, Diyi Yang
Field Experiment LLM Simulation High Quality Compost 3D Caricature

June 8, 2023

Interpretable Medical Diagnostics with Structured Data Extraction by Large Language Models
Aleksa Bisercic, Mladen Nikolic, Mihaela van der Schaar, Boris Delibasic, Pietro Lio, Andrija Petrovic
Tabular Data Clinical Text Structured Information Extraction Tabular Feature LLM Simulation

LLM Simulation

Papers

Social Science Meets LLMs: How Reliable Are Large Language Models in Social Simulations?

FPE-LLM: Highly Intelligent Time-Series Forecasting and Language Interaction LLM in Energy Systems

From a Tiny Slip to a Giant Leap: An LLM-Based Simulation for Fake News Evolution

Decoding Echo Chambers: LLM-Powered Simulations Revealing Polarization in Social Networks

adaptMLLM: Fine-Tuning Multilingual Language Models on Low-Resource Languages with Integrated LLM Playgrounds

Quantifying the Persona Effect in LLM Simulations

Systematic Biases in LLM Simulations of Debates

The Challenge of Using LLMs to Simulate Human Behavior: A Causal Inference Perspective

CoMPosT: Characterizing and Evaluating Caricature in LLM Simulations

Interpretable Medical Diagnostics with Structured Data Extraction by Large Language Models