Back to AI News
arxiv

PhyCAD: Physics-Aware Text-to-CAD Generation via Hierarchical Reinforcement Learning

PhyCAD is a framework that trains language models to produce executable CadQuery programs satisfying both geometric and physical constraints. The work introduces a Text-to-CAD dataset linking parametric CAD models with geometric specifications, loading conditions, and physical constraints, plus an automated CAD-CAE verifier combining program execution, geometric checks, and finite element analysis. It proposes a Hierarchical Balanced Advantage for Group Relative Policy Optimization that groups candidates by verification status and preserves cross-level priority while providing fine-grained feedback. Experiments across multiple language models show improved joint geometric and physical constraint satisfaction.

AI Brief

It offers a concrete research advance in AI-driven engineering design, showing how reinforcement learning with verifiable CAD-CAE feedback can move text-to-CAD beyond geometric reconstruction toward physics-aware design.

arXiv:2611.09401v1 Announce Type: new Abstract: Language-driven computer-aided design (CAD) generation enables the creation of editable parametric models from text, yet geometric correctness alone does not ensure physical feasibility. We present PhyCAD, a framework for physics-aware Text-to-CAD generation that trains language models to produce executable CadQuery programs under geometric and physical constraints. To support this setting, we construct a Text-to-CAD dataset that links parametric CAD models with geometric specifications, loading conditions, and physical constraints. We further develop an automated CAD--CAE verifier that combines program execution, geometric checks, and finite element analysis to provide feedback for model training. A key learning challenge is to prioritize joint satisfaction of geometric and physical constraints while retaining informative feedback on partial progress. A flat aggregation of heterogeneous rewards can obscure this priority, allowing improvements in individual criteria to compensate for unmet design requirements. We therefore introduce a Hierarchical Balanced Advantage for Group Relative Policy Optimization, grouping candidates into levels according to their verification status and combining between-level progress with objective-wise peer ranking within each level. This formulation preserves cross-level priority in the training signal while providing fine-grained feedback for satisfying constraints and improving feasible designs. Experiments across multiple language models show that PhyCAD consistently improves joint geometric and physical constraint satisfaction. These results demonstrate the value of verifiable engineering feedback for advancing language-driven CAD generation beyond geometric reconstruction toward physics-aware design.
Be the first to rate this article.

Comments & Questions (0)

Captcha
Please be respectful — let's keep the conversation friendly.

No comments yet

Be the first to comment!