@inproceedings{bougie-etal-2026-beyond,
title = "Beyond Offline A/{B} Testing: Context-Aware Agent Simulation for Recommender System Evaluation",
author = "Bougie, Nicolas and
Marconi, Gian Maria and
Ye, Xiaotong and
Watanabe, Narimawa",
editor = "Atanasova, Pepa and
Haffari, Gholamreza and
Wang, Minghan",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-industry.2/",
pages = "5--24",
ISBN = "979-8-89176-460-6",
abstract = "Recommender systems are central to online services, enabling users to navigate through massive amounts of content across various domains. However, their evaluation remains challenging due to the disconnect between offline metrics and online performance. The emergence of Large Language Model-powered agents offers a promising solution, yet existing studies model users in isolation, neglecting the contextual factors such as time, location, and needs, which fundamentally shape human decision-making. We introduce ContextSim, an LLM agent framework that simulates believable user proxies by anchoring interactions in daily life activities. A life simulation module generates scenarios specifying when, where, and why users engage with recommendations. To align preferences with genuine humans, we model agents' internal thoughts and enforce consistency at both action and trajectory levels. Experiments across domains demonstrate closer alignment with human behavior than prior work. We further validate our approach through offline A/B testing correlation and show that RS parameters optimized using ContextSim yield improved real-world engagement."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="bougie-etal-2026-beyond">
<titleInfo>
<title>Beyond Offline A/B Testing: Context-Aware Agent Simulation for Recommender System Evaluation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Nicolas</namePart>
<namePart type="family">Bougie</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gian</namePart>
<namePart type="given">Maria</namePart>
<namePart type="family">Marconi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaotong</namePart>
<namePart type="family">Ye</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Narimawa</namePart>
<namePart type="family">Watanabe</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track</title>
</titleInfo>
<name type="personal">
<namePart type="given">Pepa</namePart>
<namePart type="family">Atanasova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gholamreza</namePart>
<namePart type="family">Haffari</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Minghan</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-460-6</identifier>
</relatedItem>
<abstract>Recommender systems are central to online services, enabling users to navigate through massive amounts of content across various domains. However, their evaluation remains challenging due to the disconnect between offline metrics and online performance. The emergence of Large Language Model-powered agents offers a promising solution, yet existing studies model users in isolation, neglecting the contextual factors such as time, location, and needs, which fundamentally shape human decision-making. We introduce ContextSim, an LLM agent framework that simulates believable user proxies by anchoring interactions in daily life activities. A life simulation module generates scenarios specifying when, where, and why users engage with recommendations. To align preferences with genuine humans, we model agents’ internal thoughts and enforce consistency at both action and trajectory levels. Experiments across domains demonstrate closer alignment with human behavior than prior work. We further validate our approach through offline A/B testing correlation and show that RS parameters optimized using ContextSim yield improved real-world engagement.</abstract>
<identifier type="citekey">bougie-etal-2026-beyond</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-industry.2/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>5</start>
<end>24</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Beyond Offline A/B Testing: Context-Aware Agent Simulation for Recommender System Evaluation
%A Bougie, Nicolas
%A Marconi, Gian Maria
%A Ye, Xiaotong
%A Watanabe, Narimawa
%Y Atanasova, Pepa
%Y Haffari, Gholamreza
%Y Wang, Minghan
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-460-6
%F bougie-etal-2026-beyond
%X Recommender systems are central to online services, enabling users to navigate through massive amounts of content across various domains. However, their evaluation remains challenging due to the disconnect between offline metrics and online performance. The emergence of Large Language Model-powered agents offers a promising solution, yet existing studies model users in isolation, neglecting the contextual factors such as time, location, and needs, which fundamentally shape human decision-making. We introduce ContextSim, an LLM agent framework that simulates believable user proxies by anchoring interactions in daily life activities. A life simulation module generates scenarios specifying when, where, and why users engage with recommendations. To align preferences with genuine humans, we model agents’ internal thoughts and enforce consistency at both action and trajectory levels. Experiments across domains demonstrate closer alignment with human behavior than prior work. We further validate our approach through offline A/B testing correlation and show that RS parameters optimized using ContextSim yield improved real-world engagement.
%U https://aclanthology.org/2026.emnlp-industry.2/
%P 5-24
Markdown (Informal)
[Beyond Offline A/B Testing: Context-Aware Agent Simulation for Recommender System Evaluation](https://aclanthology.org/2026.emnlp-industry.2/) (Bougie et al., EMNLP 2026)
ACL