Research map / Large language models / AI and society
AIMold: An Autonomous AI-based Pipeline for Complex Mold Design
Pengyun Qiu, Shuo Wang, Zeyuan Chen, Yihao Zhi, Chongjie Ye, XIAOGUANG HAN
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 |
|---|---|---|---|
| add second-stage voxel refinement to the first-stage generator (ablation) | AIMold · MoldCAD auxiliary components | COV (%): 55.68 -> 57.29 | Improved |
| add second-stage voxel refinement to the first-stage generator (ablation) | AIMold · MoldCAD auxiliary components | MMD: 0.0140 -> 0.0147 | Hurt |
| add second-stage voxel refinement to the first-stage generator (ablation) | AIMold · MoldCAD auxiliary components | JSD: 0.054 -> 0.050 | Improved |
| add second-stage voxel refinement to the first-stage generator (ablation) | AIMold · MoldCAD upper and lower molds | COV (%): 46.55 -> 49.01 | Improved |
| add second-stage voxel refinement to the first-stage generator (ablation) | AIMold · MoldCAD upper and lower molds | MMD: 0.0096 -> 0.0091 | Improved |
| add second-stage voxel refinement to the first-stage generator (ablation) | AIMold · MoldCAD upper and lower molds | JSD: 0.073 -> 0.061 | Improved |
| use 12 render views instead of four for structural latent encoding (ablation) | AIMold · MoldCAD auxiliary components | COV (%): 55.44 -> 57.29 | Improved |
| use 12 render views instead of four for structural latent encoding (ablation) | AIMold · MoldCAD auxiliary components | MMD: 0.0146 -> 0.0147 | Hurt |
| use pretrained large-scale 3D encoder instead of training from scratch (ablation) | AIMold · MoldCAD auxiliary components | COV (%): 52.82 -> 57.29 | Improved |
| use pretrained large-scale 3D encoder instead of training from scratch (ablation) | AIMold · MoldCAD auxiliary components | MMD: 0.0173 -> 0.0147 | Improved |
| use pretrained large-scale 3D encoder instead of training from scratch (ablation) | AIMold · MoldCAD auxiliary components | JSD: 0.0682 -> 0.050 | Improved |
| replace direct orientation regression with plane-detection candidate classification | orientation estimation · MoldCAD orientation estimation | Accuracy (%): 12.66 -> 91.17 | Improved |
Closest papers by meaning
- Learning CAD Modeling Sequences via Projection and Part Awareness (NeurIPS 2025 · similarity 0.86)
- Position: You Can't Manufacture a NeRF (ICML 2025 · similarity 0.85)
- CADCrafter: Generating Computer-Aided Design Models from Unconstrained Images (CVPR 2025 · similarity 0.84)
- FlexCAD: Unified and Versatile Controllable CAD Generation with Fine-tuned Large Language Models (ICLR 2025 · similarity 0.84)
- CReFT-CAD: Boosting Orthographic Projection Reasoning for CAD via Reinforcement Fine-Tuning (NeurIPS 2025 · similarity 0.84)
- StepCAD: Mesh-to-CAD Code Generation via LLM Policy and Geometry-Guided Search (NeurIPS 2026 · similarity 0.84)
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.
