Research map / Image generation / Autoregressive and latent generation
MamTiff-CAD: Multi-Scale Latent Diffusion with Mamba+ for Complex Parametric Sequence
Liyuan Deng, Yunpeng Bai, Yongkang Dai, Xiaoshui Huang, Hongping Gan, Dongshuo Huang, Hao Jiacheng, Yilei Shi
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 Mamba+ block instead of Transformer block in the encoder (ablation) | MamTiff-CAD autoencoder · ABC-256 reconstruction | parameter accuracy: 63.62 -> 99.93 | Improved |
| Add Mamba+ block instead of Transformer block in the encoder (ablation) | MamTiff-CAD autoencoder · ABC-256 reconstruction | step ratio: 34.28% -> 93.93% | Improved |
| Add forget gate (Mamba+) to standard Mamba (ablation) | MamTiff-CAD autoencoder · ABC-256 reconstruction | step ratio: 92.01% -> 93.93% | Improved |
| Add forget gate (Mamba+) to standard Mamba (ablation) | MamTiff-CAD autoencoder · ABC-256 reconstruction | invalid ratio: 10.35% -> 8.50% | Improved |
| Add forget gate (Mamba+) to standard Mamba (ablation) | MamTiff-CAD autoencoder · ABC-256 reconstruction | MCD: 0.76 -> 0.75 | No clear effect |
| Add multi-scale transformer (MST) to the diffusion generator (ablation) | MamTiff-CAD generator · Unconditional generation | JSD: 4.92 -> 3.19 | Improved |
| Add multi-scale transformer (MST) to the diffusion generator (ablation) | MamTiff-CAD generator · Unconditional generation | step ratio: 77.05% -> 85.38% | Improved |
| Add multi-scale transformer (MST) to the diffusion generator (ablation) | MamTiff-CAD generator · Unconditional generation | MMD: 1.47 -> 1.32 | Improved |
| Add multi-scale transformer (MST) to the diffusion generator (ablation) | MamTiff-CAD generator · Unconditional generation | novelty: 99.6 -> 99.4 | No clear effect |
| Use MamTiff-CAD instead of DeepCAD for reconstruction | DeepCAD · ABC-256 reconstruction | MCD: 41.02 -> 0.75 | Improved |
| Use MamTiff-CAD instead of DeepCAD zero-shot on Fusion360 | DeepCAD · Fusion360 reconstruction | parameter accuracy: 77.99 -> 97.99 | Improved |
| Use MamTiff-CAD instead of HNC-CAD for generation | HNC-CAD · Unconditional generation | JSD: 4.25 -> 3.19 | Improved |
Closest papers by meaning
- Mamba-CAD: State Space Model for 3D Computer-Aided Design Generative Modeling (AAAI 2025 · similarity 0.90)
- CAD-Llama: Leveraging Large Language Models for Computer-Aided Design Parametric 3D Model Generation (CVPR 2025 · similarity 0.87)
- Text-to-CAD Generation Through Infusing Visual Feedback in Large Language Models (ICML 2025 · similarity 0.87)
- Draw Step by Step: Reconstructing CAD Construction Sequences from Point Clouds via Multimodal Diffusion. (CVPR 2024 · similarity 0.87)
- cadrille: Multi-modal CAD Reconstruction with Reinforcement Learning (ICLR 2026 · similarity 0.86)
- CAD-Coder: Text-to-CAD Generation with Chain-of-Thought and Geometric Reward (NeurIPS 2025 · similarity 0.85)
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.
