Reproduce “Rethinking Human Intent-to-CAD: Parametric CAD Model Generation via Cooperative Multi-Task Alignment and Spatial-Aware Reinforcement Learning” 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
Qingwang Zhang, Jiahao Li, Xiangdong Zhou
HiCAD generates executable CadQuery code from sketches, text, or both, with one model trained across four input configurations. CMTA interleaves tasks; SARL scores generated programs for validity and spatial agreement. On 100 volunteer-drawn test sketches with complete text, HiCAD reaches 73.37% IoU and 0% invalid rate, versus Cadrille’s 59.20% and 9%. This small paired evaluation does not establish engineering reliability.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.