WebVR: Benchmarking Multimodal LLMs for WebPage Recreation from Videos via Human-Aligned Visual Rubrics
This addresses the lack of dedicated benchmarks for video-to-webpage generation, which is incremental as it builds on existing web-generation methods by focusing on video inputs.
The authors tackled the problem of video-conditioned webpage generation by introducing WebVR, a benchmark with 175 synthetic webpages and a human-aligned visual rubric, revealing substantial gaps in model performance on fine-grained style and motion quality while achieving 96% agreement between automatic and human evaluations.
Existing web-generation benchmarks rely on text prompts or static screenshots as input. However, videos naturally convey richer signals such as interaction flow, transition timing, and motion continuity, which are essential for faithful webpage recreation. Despite this potential, video-conditioned webpage generation remains largely unexplored, with no dedicated benchmark for this task. To fill this gap, we introduce WebVR, a benchmark that evaluates whether MLLMs can faithfully recreate webpages from demonstration videos. WebVR contains 175 webpages across diverse categories, all constructed through a controlled synthesis pipeline rather than web crawling, ensuring varied and realistic demonstrations without overlap with existing online pages. We also design a fine-grained, human-aligned visual rubric that evaluates the generated webpages across multiple dimensions. Experiments on 19 models reveal substantial gaps in recreating fine-grained style and motion quality, while the rubric-based automatic evaluation achieves 96% agreement with human preferences. We release the dataset, evaluation toolkit, and baseline results to support future research on video-to-webpage generation.