跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2609.36777 · v2

Code4Scene: Benchmarking Coding Agents for Constructing and Editing 3D Scenes

首次提交 2026-09-29 13:49
版本更新 2026-11-17 16:28
雷达观测 2026-11-18 19:42
arXiv ↗PDF ↗

摘要

Frontier coding agents can now write and execute code that authors 3D environments, but whether they reliably understand 3D structure and precisely control scene state remains unclear. The generated 3D scene is a persistent, executable artifact: a convincing render can hide incorrect spatial relations, intersecting objects, or unintended modifications. We introduce Code4Scene, a benchmark of 170 Unreal Engine cases built from human-assembled scenes that evaluates coding agents on two complementary settings under a shared execution interface. Construction tests scene-level spatial reasoning from open-ended language specifications, where many realizations are valid; editing tests precise control of scene state, where the agent must recover the target scene from reference images while preserving everything else. Rather than scoring code or rendered views, Code4Scene evaluates the generated engine-native scene for task fulfillment, artifact integrity, and static physical validity, with edits additionally compared against withheld ground truth. Across 14 coding-agent systems on the 83-case public set, construction and editing performance are strongly correlated but not interchangeable (Spearman ρ=0.67ρ= 0.67): Claude Fable 5.1 leads construction, Gemini 3.8 Flash leads editing. Spatial Composition is the weakest construction category for every agent, while editing remains imprecise: the best Repair F1 is only 0.562, and 34.2% of edits that fully recover the target still introduce unintended changes elsewhere in the scene. These results expose a gap between plausible 3D generation and reliable spatial reasoning and state control. The project webpage is at https://simworld.org/code4scene.

作者备注:46 pages, including appendices. v2 updates the benchmark to 170 cases and revises the evaluation results

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

本站首次收录为 v2,暂无更早版本可对照。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v2 · 2026-11-17 16:28 · 最新观测版本

Code4Scene: Benchmarking Coding Agents for Constructing and Editing 3D Scenes

Frontier coding agents can now write and execute code that authors 3D environments, but whether they reliably understand 3D structure and precisely control scene state remains unclear. The generated 3D scene is a persistent, executable artifact: a convincing render can hide incorrect spatial relations, intersecting objects, or unintended modifications. We introduce Code4Scene, a benchmark of 170 Unreal Engine cases built from human-assembled scenes that evaluates coding agents on two complementary settings under a shared execution interface. Construction tests scene-level spatial reasoning from open-ended language specifications, where many realizations are valid; editing tests precise control of scene state, where the agent must recover the target scene from reference images while preserving everything else. Rather than scoring code or rendered views, Code4Scene evaluates the generated engine-native scene for task fulfillment, artifact integrity, and static physical validity, with edits additionally compared against withheld ground truth. Across 14 coding-agent systems on the 83-case public set, construction and editing performance are strongly correlated but not interchangeable (Spearman ρ=0.67ρ= 0.67): Claude Fable 5.1 leads construction, Gemini 3.8 Flash leads editing. Spatial Composition is the weakest construction category for every agent, while editing remains imprecise: the best Repair F1 is only 0.562, and 34.2% of edits that fully recover the target still introduce unintended changes elsewhere in the scene. These results expose a gap between plausible 3D generation and reliable spatial reasoning and state control. The project webpage is at https://simworld.org/code4scene.

查看此版本解读 →查看此版本原文 ↗