Reproduce “CAD-WAM: Anticipating Geometry for Text-to-CAD Generation” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Zhengyang Li, Weijie Zheng, Yuxin Cui, Sijie Wang, Wenjie Feng, Qiyu Kang
CAD-WAM generates step-by-step CAD models by training a language model to predict each modeling operation and the resulting geometric state. At inference, it executes each operation in a CAD kernel and feeds the resulting solid into the next step. On DeepCAD, its mean Chamfer distance over valid predictions was 1.142; evaluation covers sketch-and-extrude programs, not richer CAD operations.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.