2025 · Conference paper
WebUIBench: A Comprehensive Benchmark for Evaluating Multimodal Large Language Models in WebUI-to-Code
Diagnoses webpage-to-code capability through visual perception, HTML programming, image-to-HTML understanding, and screenshot reconstruction tasks. It separates underlying skills using question-answer pairs derived from real websites.
Scope and limitations
Primarily evaluates visual and HTML reconstruction skills. The paper includes a preliminary interaction study, not a full workflow or backend-equivalence benchmark. The official repository currently exposes README and images but not the evaluation scripts described in its instructions.
Why it fits this collection
Adds diagnostic evidence for screenshot-based website reconstruction, complementary to end-to-end Design2Code evaluation.
Publication and review details
- Venue
- Findings of ACL 2025
- Status
- Conference paper
- Publication / posting date
- July 2025
- Date source
- ACL proceedings publication month.
- First arXiv submission
- June 9, 2025
- arXiv
- 2506.07818
- Reviewed arXiv version
- v1
- DOI
- 10.18653/v1/2025.findings-acl.815
The published ACL record supplies the bibliography; repository citation fields remain placeholders and the dataset card displays a different arXiv identifier. The author-linked repository is retained as a project link, not advertised as a runnable code release.
Sources reviewed
Source review only; experiments were not rerun for this collection. Cite the original authors using the linked paper or BibTeX.