WebWorld：浏览器作为自我改进Web代码的世界模型

摘要

VLM驱动的网页代码自我改进存在一个结构性缺陷：提出修复的模型正是评判该修复的模型，而在该评判者视角下，视觉合理性并不能很好地代表页面是否真正可用。该循环所缺少的是一个VLM无法欺骗的对手方，而浏览器恰恰就是这样一种对手方：它是一个确定性的、可执行的模拟器，能够模拟HTML工件在用户操作下的行为，除名称外，它就是网页代码的世界模型。我们提出WebWorld，这一接口使既有的VLM能够自主地与这个“浏览器即世界模型”交互，并决定哪些交互应成为监督信号。每一轮中，VLM生成一条批评意见，规划器将其编译为类型化的交互契约；浏览器重新执行候选方案，并且仅当目标进展以及此前验证过的所有能力都得到保持时，才签发验收证书；经过认证的转换不断累积，形成一个质量棘轮，而SFT导出唯一能看到的内容正是这一棘轮。在匹配的训练条件下，WebWorld-27B在HTMLBench-400上将Raw-27B提升了5.3分，在MiniAppBench-Val上提升了14.9分，并在交互式HTML生成方面达到了Kimi-K2.6和GPT-5.4等强大前沿系统的水平。同等规模的消融实验表明，浏览器支持的准入机制带来了这一提升：没有证书时，匹配的9B模型提升几乎消失。

English

VLM-driven self-improvement of web code has a structural flaw: the model that proposes the repair is the model that judges it, and visual plausibility under that judge is a poor proxy for whether the page actually works. What the loop is missing is a counterparty the VLM cannot fool, and the browser already is that counterparty: a deterministic, executable simulator of how an HTML artifact behaves under user actions, and in everything but name a world model for web code. We present WebWorld, the interface that lets a VLM prior interact with this browser-as-world-model autonomously and decides which interactions become supervision. Each round, the VLM emits a critique that the planner compiles into a typed interaction contract; the browser re-executes the candidate and issues an acceptance certificate only when both target progress and preservation of every previously verified capability hold; certified transitions accumulate as a quality ratchet that is the only thing the SFT export ever sees. Under matched training, WebWorld-27B improves Raw-27B by 5.3 points on HTMLBench-400 and 14.9 points on MiniAppBench-Val, and reaches the level of strong frontier systems such as Kimi-K2.6 and GPT-5.4 on interactive HTML generation. Equal-size ablations show that browser-backed admission carries the gain: without the certificate, the matched 9B lift nearly disappears.