@inproceedings{bougie-etal-2026-perceptui,
title = "{P}ercept{UI}: {LLM} Agents as Human-Aligned Synthetic Users for {UI}/{UX} Evaluation",
author = "Bougie, Nicolas and
Ye, Xiaotong and
Marconi, Gian Maria 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.22/",
pages = "329--351",
ISBN = "979-8-89176-460-6",
abstract = "User interface (UI) and user experience (UX) evaluation is central to product development, yet reliable feedback still relies on recruiting human participants or running online A/B tests, making early-stage iteration slow and costly. In light of this, recent work has explored Multimodal Large Language Models as proxy evaluators. However, existing approaches either produce surface-level critiques or a judgment that reflects the model{'}s own biases rather than the genuine response of a particular user. We introduce PerceptUI, a framework for persona-conditioned UI/UX evaluation that predicts how a specific user would answer interface-related questions and produces natural-language rationales. PerceptUI is trained in two stages: (i) contrastive reflection fine-tuning distills teacher-generated rationales by extracting lessons from human decisions, and (ii) reflective prompt-evolution, which refines prompts using the model{'}s failure traces. Across multiple domains and datasets, PerceptUI improves alignment with human responses, generalizes to unseen questions and personas, and yields population-level response distributions."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="bougie-etal-2026-perceptui">
<titleInfo>
<title>PerceptUI: LLM Agents as Human-Aligned Synthetic Users for UI/UX 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">Xiaotong</namePart>
<namePart type="family">Ye</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">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>User interface (UI) and user experience (UX) evaluation is central to product development, yet reliable feedback still relies on recruiting human participants or running online A/B tests, making early-stage iteration slow and costly. In light of this, recent work has explored Multimodal Large Language Models as proxy evaluators. However, existing approaches either produce surface-level critiques or a judgment that reflects the model’s own biases rather than the genuine response of a particular user. We introduce PerceptUI, a framework for persona-conditioned UI/UX evaluation that predicts how a specific user would answer interface-related questions and produces natural-language rationales. PerceptUI is trained in two stages: (i) contrastive reflection fine-tuning distills teacher-generated rationales by extracting lessons from human decisions, and (ii) reflective prompt-evolution, which refines prompts using the model’s failure traces. Across multiple domains and datasets, PerceptUI improves alignment with human responses, generalizes to unseen questions and personas, and yields population-level response distributions.</abstract>
<identifier type="citekey">bougie-etal-2026-perceptui</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-industry.22/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>329</start>
<end>351</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T PerceptUI: LLM Agents as Human-Aligned Synthetic Users for UI/UX Evaluation
%A Bougie, Nicolas
%A Ye, Xiaotong
%A Marconi, Gian Maria
%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-perceptui
%X User interface (UI) and user experience (UX) evaluation is central to product development, yet reliable feedback still relies on recruiting human participants or running online A/B tests, making early-stage iteration slow and costly. In light of this, recent work has explored Multimodal Large Language Models as proxy evaluators. However, existing approaches either produce surface-level critiques or a judgment that reflects the model’s own biases rather than the genuine response of a particular user. We introduce PerceptUI, a framework for persona-conditioned UI/UX evaluation that predicts how a specific user would answer interface-related questions and produces natural-language rationales. PerceptUI is trained in two stages: (i) contrastive reflection fine-tuning distills teacher-generated rationales by extracting lessons from human decisions, and (ii) reflective prompt-evolution, which refines prompts using the model’s failure traces. Across multiple domains and datasets, PerceptUI improves alignment with human responses, generalizes to unseen questions and personas, and yields population-level response distributions.
%U https://aclanthology.org/2026.emnlp-industry.22/
%P 329-351
Markdown (Informal)
[PerceptUI: LLM Agents as Human-Aligned Synthetic Users for UI/UX Evaluation](https://aclanthology.org/2026.emnlp-industry.22/) (Bougie et al., EMNLP 2026)
ACL