GuideWeb: A Benchmark for Automatic In-App Guide Generation on Real-World Web UIs

TL;DR

GuideWeb introduces a benchmark for automatic generation of in-app guides on real-world web pages, addressing the challenge of maintaining guidance as website layouts evolve, and evaluates the performance of a new guide generation model.

Contribution

This work presents GuideWeb, a novel benchmark and evaluation suite for automatic in-app guide generation on real-world web UIs, including a new model called GuideWeb Agent.

Findings

GuideWeb Agent achieves 30.79% accuracy in guide target element prediction.

BLEU scores of 44.94 for intent generation and 21.34 for guide-text generation.

Existing methods perform significantly worse, indicating the task's difficulty.

Abstract

Digital Adoption Platform (DAP) provide web-based overlays that deliver operation guidance and contextual hints to help users navigate complex websites. Although modern DAP tools enable non-experts to author such guidance, maintaining these guides remains labor-intensive because website layouts and functionalities evolve continuously, which requires repeated manual updates and re-annotation. In this work, we introduce \textbf{GuideWeb}, a new benchmark for automatic in-app guide generation on real-world web UIs. GuideWeb formulates the task as producing page-level guidance by selecting \textbf{guide target elements} grounded in the webpage and generating concise guide text aligned with user intent. We also propose a comprehensive evaluation suite that jointly measures the accuracy of guide target element selection and the quality of generated intents and guide texts. Experiments show that our proposed \textbf{GuideWeb Agent} achieves \textbf{30.79\%} accuracy in guide target element prediction, while obtaining BLEU scores of \textbf{44.94} for intent generation and \textbf{21.34} for guide-text generation. Existing baselines perform substantially worse, which highlights that automatic guide generation remains challenging and that further advances are necessary before such systems can be reliably deployed in real-world settings.