Looks Right, Works Right: A Project-Level Benchmark for Multi-Screen Mobile App Generation

2026-05-29

AI Summary

该研究通过引入MobileForge基准，解决了多屏移动应用生成中的跨页导航、代码可维护性等问题，评估了现有模型在项目级生成中的表现。

Abstract

Recent multimodal large language models can convert visual designs directly into executable code, but real mobile products require multiple screenshots to become a buildable codebase with shared components and working navigation. This project-level setting exposes three limits of existing design-to-code benchmarks: they focus on single-page generation rather than complete codebases, cannot evaluate cross-page navigation, and do not measure project-wide maintainability. We introduce MobileForge, the first benchmark for project-level multi-screen mobile app generation, comprising real mobile apps, human-reviewed screens, structured page-relationship annotations, and navigation test specifications. MobileForge supports five-axis evaluation of build, navigation, visual fidelity, code maintainability, and efficiency. We also propose state-isolated navigation testing to avoid cascading failures in navigation evaluation and an anchor-referenced list-wise visual evaluation protocol to improve visual-judge reliability. Across end-to-end runs on six frontier multimodal LLMs, current models can build mobile-app projects that compile and reach the correct pages, but interactive navigation remains unreliable and visual fidelity and maintainability still lag. The benchmark and supporting materials are available at https://github.com/anoa12159-hue/mobileforge_eval.

Problem

• multi-screen mobile app
• design-to-code
• cross-page navigation
• code maintainability

Innovation

• MobileForge
• project-level benchmark
• state-isolated navigation testing
• anchor-referenced list-wise visual evaluation