Evaluating Large Language Models as Generative User Simulators for Conversational Recommendation
This work addresses the need for cost-effective synthetic users in conversational recommendation evaluation, though it is incremental as it builds on existing methods for simulation.
The authors tackled the problem of evaluating whether large language models can accurately simulate diverse human users in conversational recommendation systems, and introduced a new five-task protocol that effectively reveals deviations from human behavior and offers strategies to reduce them.
Synthetic users are cost-effective proxies for real users in the evaluation of conversational recommender systems. Large language models show promise in simulating human-like behavior, raising the question of their ability to represent a diverse population of users. We introduce a new protocol to measure the degree to which language models can accurately emulate human behavior in conversational recommendation. This protocol is comprised of five tasks, each designed to evaluate a key property that a synthetic user should exhibit: choosing which items to talk about, expressing binary preferences, expressing open-ended preferences, requesting recommendations, and giving feedback. Through evaluation of baseline simulators, we demonstrate these tasks effectively reveal deviations of language models from human behavior, and offer insights on how to reduce the deviations with model selection and prompting strategies.