[论文] 3DCodeVerse: Exploring 3D Worlds as Code

论文概要

研究领域: CV 作者: Yipeng Gao, Ziyao Zeng, Jinfa Huang, Yijiang Li, Haolin Xiong, Wang Qin, Xinlei Yu, Youheng Yao, Jie Zhang, Kun-Yu Lin, Rong Liu, Yinyuan Zhao, Yajie Zhao, Meiqi Guo, Jiebo Luo, Lei Shu, Yunhao Ge, Xinyu Hu, Laurent Itti 发布时间: 2026-10-15 arXiv: 2610.22341

中文摘要

我们引入 3DCodeVerse，这是一个使编码智能体能够大规模创建可编辑且可验证 3D 内容的研究生态系统。包含大规模高质量可执行 3D 代码数据集、用于空间推理和自动验证的编码框架，以及 3D 编码实用指南。我们构建 3DCodeVerse-2M，通过收集开源数据、转换专门表示和用编码智能体生成新程序来构建。我们开发 3DCodeVerse Harness，这是一个多智能体有向无环图框架，使用执行检查、空间测量和视觉反馈来自动评估和修订生成的程序，无需人工反馈。大量实验表明，在我们的数据集上微调显著提升了开源模型的 3D 编码能力。我们的框架在与 Claude Design（3D Object）的成对人工评估中取得了 90% 的胜率。

原文摘要

We introduce 3DCodeVerse, a research ecosystem that enables coding agents to create editable and verifiable 3D content at scale. 3DCodeVerse comprises large-scale, high-quality datasets of executable 3D code, a coding harness for spatial reasoning and automated verification, and practical guidance for 3D coding. First, we construct 3DCodeVerse-2M, a large-scale curated dataset of standalone 3D executable code. We build this corpus by collecting open-source data, converting specialized representations, and generating new programs with coding agents. Each program is packaged with the source files and associated assets required for execution in its declared environment, along with available descriptions and verification records. Second, we develop the 3DCodeVerse Harness, a multi-agent direct...

#论文 #arXiv #CV #小凯