LandingAgent：用于落地页的参考标注数据集与智能体生成框架
LandingAgent: A Reference-Annotated Dataset and Agentic Generation Framework for Landing Pages
浏览论文内容
中文总结 AI 辅助
针对落地页生成中通用模板与无依据主张的问题，本文提出智能体框架LandingAgent，结合参考标注数据集LandingBench，实验验证其在目标适配性、呈现质量等方面优于直接生成方法。
中文摘要 AI 辅助
落地页是面向目标的网页界面，需传递针对特定目标的价值主张，同时组织信息流、视觉层级与行动号召（CTA）。尽管大语言模型可根据自然语言提示生成合理的网页代码，但直接生成常产生通用模板与无依据的说服性主张。本文研究基于目标、参考引导的落地页生成，要求系统通过适配真实页面的可复用模式（而非复制）为新目标创建可执行页面。我们引入LandingBench，这一参考-概况数据集将真实落地页抽象为区块序列、布局模式、语气描述符、视觉重点与CTA结构。基于LandingBench，我们提出LandingAgent，这一包含三个阶段的智能体框架：对目标进行概况分析、构建参考引导的线框图、通过批评引导的优化完善页面。我们在忠实度、简洁性、可读性、美观度与结构多样性方面将LandingAgent与直接提示生成进行对比评估。实验表明，LandingAgent在目标适配性、呈现质量与布局多样性上均有提升。代码可在指定URL获取。
英文摘要
Landing pages are goal-oriented web interfaces that must communicate a target-specific value proposition while organizing information flow, visual hierarchy, and calls to action (CTA). Although large language models can generate plausible webpage code from natural-language prompts, direct generation often yields generic templates and unsupported persuasive claims. We study target-grounded, reference-guided landing-page generation, where a system must create an executable page for a new target by adapting reusable patterns from real pages without copying them. We introduce LandingBench, a reference-profile dataset that abstracts real landing pages into section sequences, layout patterns, tone descriptors, visual emphasis, and CTA structure. Building on LandingBench, we propose LandingAgent, a three-phase agentic framework that profiles the target, constructs a reference-guided wireframe, and refines the page through critique-guided polishing. We evaluate LandingAgent against direct prompting on faithfulness, conciseness, readability, aesthetics, and structural diversity. Experiments show improved target grounding, presentation quality, and layout diversity. Code is available at https://github.com/IAURAI/LandingAgent.
发表机构
- Seoul National University（首尔大学）
- Samsung Electronics（三星电子）
机构由 AI 辅助整理，请以论文原文为准。
