把物理世界写成代码，AI 第一次能算准
现在的 AI 看视频能说出「球在滚」，但你要它算「这球 3 秒后到哪」，它就瞎了——因为它脑子里没有物理参数，只有画面联想。这篇换了个思路：不训练模型去「猜物理」，而是让 AI 把观察到的世界直接写成一段可运行的代码——物体的位置、速度、受力、运动方程全变成变量和公式，跑一遍代码就能算出结果。研究者还搭了个自动循环：AI 先写代码、运行、渲染成画面、跟真实视频比对、不对就改，直到代码能精确复现现实。用这套方法训练出的模型，在定量物理推理测试上超过了 GPT-4 等闭源旗舰。它不是你明天能用上的工具，但它指了个方向：与其让 AI 更会「看」，不如让它更会「算」。
📄 原文摘要(英文)
Physical understanding and reasoning depend on forming compact and generalizable representations of the world. While modern vision-language models can recognize and explain diverse physical events, they often lack explicit representations of the underlying mechanisms-such as object states, physical parameters, and governing dynamics-needed for reliably reasoning how the world evolves and responds to interventions. In this work, we introduce Code-as-World, a paradigm that represents physical worlds through executable world representations. By expressing physical composition, dynamic evolution, and visual appearance as executable code, Code-as-World provides a compact, quantitatively grounded, and controllable abstraction of the physical world. To construct such representations from multimodal observations, such as natural-language descriptions or real-world videos, we develop an agentic discovery loop inspired by abductive reasoning, where an agent proposes, executes, renders, verifies, and iteratively refines executable world hypotheses. As a concrete application, we use verified executable worlds to provide scalable physical supervision for training vision-language models on quantitative physical reasoning. Experiments show that Code-as-World-VL achieves state-of-the-art performance on QuantiPhy and surpasses leading proprietary models, highlighting the potential of executable world representations as a scalable foundation for physical intelligence.