FlowEval: Reference-based Evaluation of Generated User Interfaces

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wu, Jason, Vaithilingam, Priyan, Schoop, Eldon, Nichols, Jeffrey, Barik, Titus
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909014846078976
author Wu, Jason
Vaithilingam, Priyan
Schoop, Eldon
Nichols, Jeffrey
Barik, Titus
author_facet Wu, Jason
Vaithilingam, Priyan
Schoop, Eldon
Nichols, Jeffrey
Barik, Titus
contents While large language models (LLMs) and coding agents are often applied to user interface (UI) development, developers find it difficult to reliably assess their proficiency in visual and interaction design. Existing evaluations either rely on human experts, who can accurately assess usability by testing critical flows but are slow and costly, or on automated judges, which are scalable but less accurate and opaque. We present FlowEval, a reference-based framework that measures whether a generated UI supports realistic interaction flows by comparing navigation traces from real websites to traces from generated analogs using reference-based similarity metrics (e.g., dynamic time warping). In a small-scale study with expert UI evaluators, we show that reference-based metrics strongly correlate with human judgments, suggesting that they can provide scalable yet trustworthy evaluation for UI generation systems.
format Preprint
id arxiv_https___arxiv_org_abs_2605_04165
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle FlowEval: Reference-based Evaluation of Generated User Interfaces
Wu, Jason
Vaithilingam, Priyan
Schoop, Eldon
Nichols, Jeffrey
Barik, Titus
Multiagent Systems
Human-Computer Interaction
While large language models (LLMs) and coding agents are often applied to user interface (UI) development, developers find it difficult to reliably assess their proficiency in visual and interaction design. Existing evaluations either rely on human experts, who can accurately assess usability by testing critical flows but are slow and costly, or on automated judges, which are scalable but less accurate and opaque. We present FlowEval, a reference-based framework that measures whether a generated UI supports realistic interaction flows by comparing navigation traces from real websites to traces from generated analogs using reference-based similarity metrics (e.g., dynamic time warping). In a small-scale study with expert UI evaluators, we show that reference-based metrics strongly correlate with human judgments, suggesting that they can provide scalable yet trustworthy evaluation for UI generation systems.
title FlowEval: Reference-based Evaluation of Generated User Interfaces
topic Multiagent Systems
Human-Computer Interaction
url https://arxiv.org/abs/2605.04165