Research map / Vision-language and multimodal / Human and embodied modeling
Rethinking Human Intent-to-CAD: Parametric CAD Model Generation via Cooperative Multi-Task Alignment and Spatial-Aware Reinforcement Learning
Qingwang Zhang, Jiahao Li, Xiangdong Zhou
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace single-task SFT with cooperative multi-task alignment (CMTA) (ablation) | HiCAD · HDS & PTD-to-CAD | IoU (%): 66.61 -> 72.39 | Improved |
| replace single-task SFT with CMTA (ablation) | HiCAD · HDS-to-CAD | IoU (%): 63.84 -> 66.87 | Improved |
| replace single-task SFT with CMTA (ablation) | HiCAD · CTD-to-CAD | IoU (%): 71.01 -> 71.90 | Improved |
| replace single-task SFT with CMTA (ablation) | HiCAD · HDS & CTD-to-CAD | IoU (%): 74.35 -> 79.28 | Improved |
| add spatial-aware reinforcement learning (SARL) after CMTA (ablation) | HiCAD · HDS-to-CAD | IoU (%): 66.87 -> 69.09 | Improved |
| add SARL after CMTA (ablation) | HiCAD · HDS-to-CAD | invalid rate (%): 1.20 -> 0.22 | Improved |
| add SARL after CMTA (ablation) | HiCAD · HDS & PTD-to-CAD | IoU (%): 72.39 -> 76.03 | Improved |
| add SARL after CMTA (ablation) | HiCAD · HDS & CTD-to-CAD | IoU (%): 79.28 -> 80.90 | Improved |
| add SARL after CMTA (ablation) | HiCAD · CTD-to-CAD | median Chamfer distance: 0.29 -> 0.30 | No clear effect |
| train SARL jointly on all tasks instead of single-task SARL (ablation) | HiCAD · four CAD tasks | all metrics | Improved |
| use HiCAD instead of Cadrille | Cadrille · HDS-to-CAD invalid rate | invalid rate (%): 9.41 -> 0.37 | Improved |
| add complete text description to sketch input on real human sketches | HiCAD · Real human sketches | IoU (%): 43.27 -> 73.37 | Improved |
Closest papers by meaning
- From Intent to Execution: Multimodal Chain-of-Thought Reinforcement Learning for Precise CAD Code Generation (AAAI 2026 · similarity 0.90)
- cadrille: Multi-modal CAD Reconstruction with Reinforcement Learning (ICLR 2026 · similarity 0.88)
- FlexCAD: Unified and Versatile Controllable CAD Generation with Fine-tuned Large Language Models (ICLR 2025 · similarity 0.87)
- CAD-Coder: Text-to-CAD Generation with Chain-of-Thought and Geometric Reward (NeurIPS 2025 · similarity 0.87)
- VideoCAD: A Dataset and Model for Learning Long‑Horizon 3D CAD UI Interactions from Video (NeurIPS 2025 · similarity 0.87)
- ReCAD: Reinforcement Learning Enhanced Parametric CAD Model Generation with Vision-Language Models (AAAI 2026 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
