On LLM Wizards: Identifying Large Language Models' Behaviors for Wizard of Oz Experiments

TL;DR

This paper explores how large language models can serve as virtual Wizards in Wizard of Oz experiments, providing a systematic evaluation framework and guidelines for responsible application.

Contribution

It introduces a novel experiment lifecycle and a heuristic-based evaluation framework for assessing LLMs' role-playing abilities in WoZ studies.

Findings

LLMs can effectively role-play in WoZ experiments.

The evaluation framework reveals diverse LLM behavior patterns.

Guidelines for safe and responsible use of LLMs in research settings.

Abstract

The Wizard of Oz (WoZ) method is a widely adopted research approach where a human Wizard ``role-plays'' a not readily available technology and interacts with participants to elicit user behaviors and probe the design space. With the growing ability for modern large language models (LLMs) to role-play, one can apply LLMs as Wizards in WoZ experiments with better scalability and lower cost than the traditional approach. However, methodological guidance on responsibly applying LLMs in WoZ experiments and a systematic evaluation of LLMs' role-playing ability are lacking. Through two LLM-powered WoZ studies, we take the first step towards identifying an experiment lifecycle for researchers to safely integrate LLMs into WoZ experiments and interpret data generated from settings that involve Wizards role-played by LLMs. We also contribute a heuristic-based evaluation framework that allows the estimation of LLMs' role-playing ability in WoZ experiments and reveals LLMs' behavior patterns at scale.