Paper Detail
Juyong Jiang, Chenglin Cai, Chansung Park, Jiasi Shen, Sunghun Kim, Jianguo Li, Yue Wang
While Large Language Models (LLMs) excel at function-level code generation, project-level tasks such as generating functional and visually aesthetic multi-page websites remain highly challenging. Existing works are often limited to single-page static websites, while agentic frameworks typically rely on multi-turn execution with proprietary models, leading to substantial token costs, high latency, and brittle integration. Training a small LLM end-to-end with reinforcement learning (RL) is a promising alternative, yet it faces a critical bottleneck in designing reliable and computationally feasible rewards for website generation. Unlike single-file coding tasks that can be verified by unit tests, website generation requires evaluating inherently subjective aesthetics, cross-page interactions, and functional correctness. To this end, we propose WebGen-R1, an end-to-end RL framework tailored for project-level website generation. We first introduce a scaffold-driven structured generation paradigm that constrains the large open-ended action space and preserves architectural integrity. We then design a novel cascaded multimodal reward that seamlessly couples structural guarantees with execution-grounded functional feedback and vision-based aesthetic supervision. Extensive experiments demonstrate that our WebGen-R1 substantially transforms a 7B base model from generating nearly nonfunctional websites into producing deployable, aesthetically aligned multi-page websites. Remarkably, our WebGen-R1 not only consistently outperforms heavily scaled open-source models (up to 72B), but also rivals the state-of-the-art DeepSeek-R1 (671B) in functional success, while substantially exceeding it in valid rendering and aesthetic alignment. These results position WebGen-R1 as a viable path for scaling small open models from function-level code generation to project-level web application generation.
No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.
No ranking explanation is available yet.
No tags.
@misc{jiang2026webgen,
title = {WebGen-R1: Incentivizing Large Language Models to Generate Functional and Aesthetic Websites with Reinforcement Learning},
author = {Juyong Jiang and Chenglin Cai and Chansung Park and Jiasi Shen and Sunghun Kim and Jianguo Li and Yue Wang},
year = {2026},
abstract = {While Large Language Models (LLMs) excel at function-level code generation, project-level tasks such as generating functional and visually aesthetic multi-page websites remain highly challenging. Existing works are often limited to single-page static websites, while agentic frameworks typically rely on multi-turn execution with proprietary models, leading to substantial token costs, high latency, and brittle integration. Training a small LLM end-to-end with reinforcement learning (RL) is a promi},
url = {https://huggingface.co/papers/2604.20398},
keywords = {Large Language Models, reinforcement learning, website generation, structured generation paradigm, cascaded multimodal reward, functional correctness, aesthetic supervision, end-to-end RL framework, multi-page websites, agent-based frameworks, huggingface daily},
eprint = {2604.20398},
archiveprefix = {arXiv},
}
{}