Research

Paper

AI LLM February 27, 2026

UXSim: Towards a Hybrid User Search Simulation

Authors

Saber Zerhoudi, Michael Granitzer

Abstract

Simulating nuanced user experiences within complex interactive search systems poses distinct challenge for traditional methodologies, which often rely on static user proxies or, more recently, on standalone large language model (LLM) agents that may lack deep, verifiable grounding. The true dynamism and personalization inherent in human-computer interaction demand a more integrated approach. This work introduces UXSim, a novel framework that integrates both approaches. It leverages grounded data from traditional simulators to inform and constrain the reasoning of an adaptive LLM agent. This synthesis enables more accurate and dynamic simulations of user behavior while also providing a pathway for the explainable validation of the underlying cognitive processes.

Metadata

arXiv ID: 2602.24241
Provider: ARXIV
Primary Category: cs.IR
Published: 2026-02-27
Fetched: 2026-03-02 06:04

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2602.24241v1</id>\n    <title>UXSim: Towards a Hybrid User Search Simulation</title>\n    <updated>2026-02-27T18:14:34Z</updated>\n    <link href='https://arxiv.org/abs/2602.24241v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2602.24241v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Simulating nuanced user experiences within complex interactive search systems poses distinct challenge for traditional methodologies, which often rely on static user proxies or, more recently, on standalone large language model (LLM) agents that may lack deep, verifiable grounding. The true dynamism and personalization inherent in human-computer interaction demand a more integrated approach. This work introduces UXSim, a novel framework that integrates both approaches. It leverages grounded data from traditional simulators to inform and constrain the reasoning of an adaptive LLM agent. This synthesis enables more accurate and dynamic simulations of user behavior while also providing a pathway for the explainable validation of the underlying cognitive processes.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.IR'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.HC'/>\n    <published>2026-02-27T18:14:34Z</published>\n    <arxiv:primary_category term='cs.IR'/>\n    <arxiv:journal_ref>Proceedings of the 34th ACM International Conference on Information and Knowledge Management (CIKM '25), November 10--14, 2025, Seoul, Republic of Korea</arxiv:journal_ref>\n    <author>\n      <name>Saber Zerhoudi</name>\n    </author>\n    <author>\n      <name>Michael Granitzer</name>\n    </author>\n    <arxiv:doi>10.1145/3746252.3761640</arxiv:doi>\n    <link href='https://doi.org/10.1145/3746252.3761640' rel='related' title='doi'/>\n  </entry>"
}