Generative Agents Navigating Digital Libraries

TL;DR

This paper introduces Agent4DL, a user search behavior simulator for digital libraries that generates realistic, diverse, and context-aware user interactions, validated against real user data and outperforming existing simulators.

Contribution

We present Agent4DL, a novel simulator that models realistic user search behaviors in digital libraries, addressing data scarcity and enhancing simulation accuracy.

Findings

Agent4DL accurately mimics real user interactions.

It outperforms existing simulators like SimIIR 2.0.

Generates diverse, context-aware user behaviors.

Abstract

In the rapidly evolving field of digital libraries, the development of large language models (LLMs) has opened up new possibilities for simulating user behavior. This innovation addresses the longstanding challenge in digital library research: the scarcity of publicly available datasets on user search patterns due to privacy concerns. In this context, we introduce Agent4DL, a user search behavior simulator specifically designed for digital library environments. Agent4DL generates realistic user profiles and dynamic search sessions that closely mimic actual search strategies, including querying, clicking, and stopping behaviors tailored to specific user profiles. Our simulator's accuracy in replicating real user interactions has been validated through comparisons with real user data. Notably, Agent4DL demonstrates competitive performance compared to existing user search simulators such as SimIIR 2.0, particularly in its ability to generate more diverse and context-aware user behaviors.