MV-Bench: Benchmarking Multimodal Large Language Models for Coordinated Multi-View Interface Construction
- ID: 163211b7
- 原文链接: https://arxiv.org/abs/2607.19910
- PDF: https://arxiv.org/pdf/2607.19910
- 作者: Yue Zhao, Hongxu Liu, Feiyu Wang, Xiaoyu Yang, Tong Ge, Zhen Yang, Chao Wang, Qiong Zeng
- 日期: 2026-07-22
- 分类: cs.CV, cs.HC
- 标签: multimodal-llm, benchmark, visualization, ui-generation, evaluation
- 质量评分: 4/5
- 抓取时间: 2026-07-24T12:19:13+08:00
中文解读
MV-Bench 把多模态大模型的 UI/可视化生成评测从单图表推进到 coordinated multi-view interface：用 Tableau workbook 作为 ground truth，因为其中显式包含数据绑定视觉映射和交互基准包含 92 个基础界面1,048 个验证实例摘要显示最强模型视觉布局可达 75.45%，但数据绑定只有 21.71%交互完整性 11.68%，说明看起来像距离可用交互界面仍很远
为什么值得关注
- 把多模态生成评测推进到含数据绑定与交互逻辑的多视图界面，暴露“视觉像”与“真实可用”之间的能力缺口。
English Summary
MV-Bench evaluates MLLMs on coordinated multi-view interface construction rather than single-chart generation. It uses Tableau workbooks as ground truth because they encode data bindings, visual mappings, and interactions, then converts them into executable web interfaces. The benchmark has 92 base interfaces and 1,048 verified instances. In a single-pass setting, the strongest tested model reached 75.45% visual layout accuracy but only 21.71% data binding correctness and 11.68% interaction completeness.
Abstract
Multimodal large language models (MLLMs) are increasingly expected to automate visualization development by generating code directly from visual designs. However, existing evaluations mainly focus on single-chart generation and overlook coordinated multi-view interface construction, which requires joint reasoning about data semantics, view coordination, and interaction logic. Consequently, MLLM capabilities in this setting remain underexplored, and the field lacks a dedicated benchmark for systematic assessment. We introduce MV-Bench, a benchmark for evaluating MLLMs on coordinated multi-view interface construction. Instead of relying on incomplete or inconsistent open-source implementations, we use Tableau workbook files as ground truth because they explicitly encode data bindings, visual mappings, and interactions. We develop a multi-stage pipeline that converts these specifications into executable web interfaces through structured intermediate representations. The benchmark contains 92 base interfaces and 1,048 verified instances created by recombining chart types, datasets, and interaction patterns. Each instance includes executable code, a rendered interface, a dataset, and interaction annotations. We evaluate five state-of-the-art MLLMs in a single-pass setting using metrics for visual fidelity, data binding correctness, and interaction completeness. The strongest model achieves 75.45 percent accuracy in visual layout reproduction, but only 21.71 percent in data binding and 11.68 percent in interaction completeness. These results show that current MLLMs can reproduce visual appearance but remain limited in generating the data semantics and interactive logic required by coordinated multi-view interfaces. Iterative refinement improves code executability but does not substantially reduce the gap in data binding and interaction generation.
Metadata
- arXiv ID: 2607.19910
- Primary category: cs.CV
- Categories: cs.CV, cs.HC
- Source: OpenCLI arXiv metadata