首个面向工程建造的物理对齐交互评测基准，评估大模型按语言指令构建真实结构的能力。
BuildArena: A Physics-Aligned Interactive Benchmark of LLMs for Engineering Construction
- 设计多难度层级任务，覆盖静力学与动力学场景，支持可扩展评估。
- 在9个前沿大模型上测试，验证其基于语言指令生成符合物理规律结构的能力。
- 适合关注建筑自动化、多模态推理与物理对齐的AI研究者使用。
工程建造自动化旨在将自然语言描述转化为物理可行的结构，需在严格物理约束下进行复杂综合推理。尽管现代大模型具备广泛知识和强大推理能力，其在建造任务中的实际能力仍缺乏系统评估。为此，我们提出BuildArena，首个面向语言驱动工程建造的物理对齐交互评测基准。技术上，该基准在两方面贡献：(1) 可扩展的任务设计策略，涵盖静态与动态力学，分多个难度层级；(2) 3D空间几何计算库，支持基于语言指令的结构构建。我们在9个前沿大模型及3个开源模型上全面评估其语言驱动、物理约束下的建造自动化能力。代码已开源（https://github.com/AI4Science-WestlakeU/BuildArena），助力工程自动化发展。
原文摘要 · Abstract (English)
Engineering construction automation aims to transform natural language specifications into physically viable structures, requiring complex integrated reasoning under strict physical constraints. While modern LLMs possess broad knowledge and strong reasoning capabilities that make them promising candidates for this domain, their construction competencies remain largely unevaluated. To address this gap, we introduce BuildArena, the first physics-aligned interactive benchmark designed for language-driven engineering construction. Technically, it contributes to the community in two aspects: (1) an extendable task design strategy spanning static and dynamic mechanics across multiple difficulty tiers; (2) a 3D Spatial Geometric Computation Library for supporting construction based on language instructions. On nine frontier LLMs and three additional open-weight models, BuildArena comprehensively evaluates their capabilities for language-driven and physics-grounded construction automation. We release the code at https://github.com/AI4Science-WestlakeU/BuildArena to benefit construction automation in engineering applications.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。