WebRenderBench: Enhancing Web Interface Generation through Layout-Style Consistency and Reinforcement Learning
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916997745344512 |
|---|---|
| author | Lai, Peichao Zhuang, Jinhui Zhang, Kexuan Xiong, Ningchang Wang, Shengjie Xu, Yanwei Chen, Chong Wang, Yilei Cui, Bin |
| author_facet | Lai, Peichao Zhuang, Jinhui Zhang, Kexuan Xiong, Ningchang Wang, Shengjie Xu, Yanwei Chen, Chong Wang, Yilei Cui, Bin |
| contents | Automating the conversion of UI images into web code is a critical task for front-end development and rapid prototyping. Advances in multimodal large language models (MLLMs) have made WebUI-to-Code increasingly feasible, yet existing benchmarks remain limited in data diversity and evaluation reliability. To address these issues, we present WebRenderBench, a large-scale benchmark of 45.1k webpages collected from real-world portal sites, offering greater diversity, complexity, and realism than prior benchmarks. We further propose a novel evaluation metric that measures layout and style consistency from the final rendered pages. Unlike vision-based methods that rely on costly LLM reasoning or structure-based comparisons vulnerable to noise and asymmetry, our approach enables more efficient, objective, and reliable UI quality assessment. Finally, we introduce the Automated Layout and Style Inspection Agent (ALISA), which integrates this metric into reinforcement learning as a reward signal to enhance training on crawled asymmetric webpages. Experiments show that ALISA significantly boosts generation performance, achieving state-of-the-art results across multiple metrics. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_04097 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | WebRenderBench: Enhancing Web Interface Generation through Layout-Style Consistency and Reinforcement Learning Lai, Peichao Zhuang, Jinhui Zhang, Kexuan Xiong, Ningchang Wang, Shengjie Xu, Yanwei Chen, Chong Wang, Yilei Cui, Bin Artificial Intelligence Automating the conversion of UI images into web code is a critical task for front-end development and rapid prototyping. Advances in multimodal large language models (MLLMs) have made WebUI-to-Code increasingly feasible, yet existing benchmarks remain limited in data diversity and evaluation reliability. To address these issues, we present WebRenderBench, a large-scale benchmark of 45.1k webpages collected from real-world portal sites, offering greater diversity, complexity, and realism than prior benchmarks. We further propose a novel evaluation metric that measures layout and style consistency from the final rendered pages. Unlike vision-based methods that rely on costly LLM reasoning or structure-based comparisons vulnerable to noise and asymmetry, our approach enables more efficient, objective, and reliable UI quality assessment. Finally, we introduce the Automated Layout and Style Inspection Agent (ALISA), which integrates this metric into reinforcement learning as a reward signal to enhance training on crawled asymmetric webpages. Experiments show that ALISA significantly boosts generation performance, achieving state-of-the-art results across multiple metrics. |
| title | WebRenderBench: Enhancing Web Interface Generation through Layout-Style Consistency and Reinforcement Learning |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2510.04097 |