BuildArena: A Physics-Aligned Interactive Benchmark of LLMs for Engineering Construction

Tian Xia, Tianrun Gao, Wenhao Deng, Long Wei, Xiaowei Qian, Chenglei Yu, Tailin Wu
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:138165-138200, 2026.

Abstract

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.

Cite this Paper


BibTeX
@InProceedings{pmlr-v306-xia26d, title = {{B}uild{A}rena: A Physics-Aligned Interactive Benchmark of {LLM}s for Engineering Construction}, author = {Xia, Tian and Gao, Tianrun and Deng, Wenhao and Wei, Long and Qian, Xiaowei and Yu, Chenglei and Wu, Tailin}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {138165--138200}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/xia26d/xia26d.pdf}, url = {https://proceedings.mlr.press/v306/xia26d.html}, abstract = {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.} }
Endnote
%0 Conference Paper %T BuildArena: A Physics-Aligned Interactive Benchmark of LLMs for Engineering Construction %A Tian Xia %A Tianrun Gao %A Wenhao Deng %A Long Wei %A Xiaowei Qian %A Chenglei Yu %A Tailin Wu %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-xia26d %I PMLR %P 138165--138200 %U https://proceedings.mlr.press/v306/xia26d.html %V 306 %X 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.
APA
Xia, T., Gao, T., Deng, W., Wei, L., Qian, X., Yu, C. & Wu, T.. (2026). BuildArena: A Physics-Aligned Interactive Benchmark of LLMs for Engineering Construction. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:138165-138200 Available from https://proceedings.mlr.press/v306/xia26d.html.

Related Material