Mamba-CAD: State Space Model For 3D Computer-Aided Design Generative Modeling

Xueyang Li; Yunzhong Lou; Yu Song; Xiangdong Zhou

cs.CVFeb 28, 2026

Mamba-CAD: State Space Model For 3D Computer-Aided Design Generative Modeling

Authors: Xueyang Li, Yunzhong Lou, Yu Song, Xiangdong Zhou

Organizations: School of Computer Science, Fudan University, Shanghai, China

Abstract

Computer-Aided Design (CAD) generative modeling has a strong and long-term application in the industry. Recently, the parametric CAD sequence as the design logic of an object has been widely mined by sequence models. However, the industrial CAD models, especially in component objects, are fine-grained and complex, requiring a longer parametric CAD sequence to define. To address the problem, we introduce Mamba-CAD, a self-supervised generative modeling for complex CAD models in the industry, which can model on a longer parametric CAD sequence. Specifically, we first design an encoder-decoder framework based on a Mamba architecture and pair it with a CAD reconstruction task for pre-training to model the latent representation of CAD models; and then we utilize the learned representation to guide a generative adversarial network to produce the fake representation of CAD models, which would be finally recovered into parametric CAD sequences via the decoder of MambaCAD. To train Mamba-CAD, we further create a new dataset consisting of 77,078 CAD models with longer parametric CAD sequences. Comprehensive experiments are conducted to demonstrate the effectiveness of our model under various evaluation metrics, especially in the generation length of valid parametric CAD sequences. The code and dataset can be achieved from https://github.com/Sunny-Hack/Code-for-Mamba-CAD-AAAI-2025-.

Figures & tables

Figure 1: The gallery of CAD models created by Mamba-CAD, including visual illustrations of the created CAD models from CAD sequence reconstruction, completion, and random generation.

Figure 2: Common CAD commands discussed in this paper.

<

SOL

>

and

<

EOS

>

are two symbols for representing the start of one loop and the end of a whole parametric CAD sequence, respectively.

f

is a counter-clockwise flag.

Figure 3: The overview of our Mamba-CAD architecture. The whole pipeline can be divided into three parts which marked in three different colored line-arrows. (Pre-training stage $\longrightarrow$ ) : We leverage a task of CAD sequence reconstruction to pre-train the encoder and decoder of Mamba-CAD. Given a mini-batch of CAD-models, their corresponding parametric CAD sequences would be first discretized into friendly network representations (Sec. CAD sequence representation CAD sequence representation ). Next, these representations keep going forward Fusion Embedding, Encoder, and Decoder of Mamba-CAD (Sec. Mamba-CAD Mamba-CAD ) to recover predicted parametric CAD sequences. ( Training stage $\longrightarrow$ ) : When the pre-training stage is complete, the encoder parameters of Mamba-CAD would be frozen to produce the latent representation $\mathbb{K}$ that is further used to guide a 1-D Latent-GAN training. ( Generation stage $\longrightarrow$ ) : Once the 1-D Latent GAN is well-trained, it enables to randomly sample the noise from a standard gaussian distribution and further feed it into Generator to produce the fake latent representation $\mathbb{F}$ , which would further undergo the decoder of Mamba-CAD to recover the parametric CAD sequence.

Dataset	TN	AL	[1-10]	[11-25]	[26-40]	[41-60]	[60-128]
DeepCAD	178,238	15	44.58%	38.73%	11.38%	5.31%	-
Mamba-CAD	77,078	38	-	41.25%	26.60%	16.13%	16.02%

Table 1: The statistic comparison of between DeepCAD dataset and Our Mamba-CAD dataset, including total number (TN), average length (AL) of parametric CAD sequences, and the length distribution.

Figure 4: Visual illustrations of CAD sequence reconstruction. GT represents Ground Truth. $\emptyset$ denotes the CAD sequence is invalid, leading to the failure of constructing a 3D shape. Please enlarge 200% for a clear view.

Method	$A_{c}\uparrow$	$A_{p}\uparrow$	MCD $\downarrow$	IR $\downarrow$	AL $\uparrow$	L $\geq$ 60 $\uparrow$	SR $\uparrow$
DeepCAD	96.49	87.38	2.97	14.83%	32.52	51.82%	86.58%
DeepCAD (++)	96.92	87.68	2.81	14.05%	32.49	50.85%	87.49%
TM-CAD	95.73	85.48	4.82	15.85%	31.21	50.31%	85.61%
Mamba-CAD	99.96	99.93	0.64	5.41%	38.91	85.18%	95.43%

Table 2: The comparison of Mamba-CAD and variants in CAD sequence reconstruction. MCD is multiplied by

10^{3}

A_{c}

A_{p}

, and IR are multiplied by 100%. L

\geq

60: the ratio of valid reconstructed CAD sequences (Length

\geq

60) in total CAD sequences (Length

\geq

60) within test set.

Method	$A_{c}\uparrow$	$A_{p}\uparrow$	MCD $\downarrow$	IR $\downarrow$	AL $\uparrow$	L $\geq$ 60 $\uparrow$	SR $\uparrow$
DeepCAD	83.75	69.77	79.08	42.36%	28.72	19.11%	59.19%
DeepCAD (++)	85.95	71.27	69.68	40.19%	29.78	20.74%	61.12%
TM-CAD	80.75	67.71	82.34	44.34%	28.26	17.79%	57.23%
HNC-CAD	-	-	-	-	30.51	42.89%	74.36%
Mamba-CAD	99.85	98.91	1.03	8.03%	35.43	79.46%	92.84%

Table 3: Quantitative results of CAD sequence completion.

Method	$A_{c}\uparrow$	$A_{p}\uparrow$	MCD $\downarrow$	IR $\downarrow$	SR $\uparrow$
DeepCAD	95.23	85.58	10.97	10.78%	90.34%
DeepCAD (++)	96.79	87.72	8.93	9.35%	91.22%
TM-CAD	90.94	83.85	12.46	12.37%	87.61%
Mamba-CAD	99.21	98.21	1.31	3.72%	96.51%

Table 4: The quantitative results of generalization ability in Fusion360 reconstruction dataset.

Figure 5: Visualizations of random generation of parametric CAD sequences.

Method	MMD $\downarrow$	JSD $\downarrow$	COV $\uparrow$	Unique $\uparrow$	Novel $\uparrow$	AL $\uparrow$	SR $\uparrow$
DeepCAD	3.51	8.93	63.77%	83.41	51.94	27.78	62.76%
SkexGen	2.96	7.39	65.89%	93.33	60.47	29.18	63.49%
HNC-CAD	2.65	6.07	70.43%	92.18	55.32	31.36	68.84%
Mamba-CAD	1.94	4.28	75.61%	94.73	63.25	36.75	74.13%

Table 5: The quantitative results of random generation of parametric CAD sequences. MMD, JSD, Unique, and Novel are all multiplied by

10^{2}

Method	$A_{c}\uparrow$	$A_{p}\uparrow$	MCD $\downarrow$	IR $\downarrow$	AL $\uparrow$	L $\geq$ 60 $\uparrow$	SR $\uparrow$
A	93.67	83.41	7.89	20.79%	26.93	43.57%	79.36%
B	94.23	83.95	5.67	18.04%	30.48	50.26%	82.77%
C	99.91	99.57	0.69	6.29%	38.16	84.53%	93.91%
Mamba-CAD	99.96	99.93	0.64	5.41%	38.91	85.18%	95.43%

Table 6: Ablation studies of Mamba-CAD.

Parametric CAD Modeling

Parametric CAD modeling represents 3D designs as editable, constraint-driven construction histories and develops methods to generate, reconstruct, and revise them.

See all 81 papers

GeoFusion-CAD: Structure-Aware Diffusion with Geometric State Space for Parametric 3D Design
Mar 23, 2026Xiaolei Zhou, Chuangjie Fang, Jie Wu +3Parametric CAD Modeling 3D Diffusion Models
Zhejiang University of Technology · Hangzhou International Innovation Institute, Beihang University
DreamCAD: Scaling Multi-modal CAD Generation using Differentiable Parametric Surfaces
Mar 5, 2026Mohammad Sadil Khan, Muhammad Usama, Rolandos Alexandros Potamias +4Text-to-CAD Generation Parametric CAD Modeling
DFKI, Germany · RPTU Kaiserslautern, Germany · Imperial College London, United Kingdom +1
GuideCAD: A Lightweight Multimodal Framework for 3D CAD Model Generation via Prefix Embedding
Jun 5, 2026Minseong Kim, Jinyeong Park, Sungho Park +13D Asset Generation Parametric CAD Modeling
Convergence Research Center for Insect Vectors Incheon, South Korea · Department of Computer Science and Engineering Incheon National University Incheon, 22012, South Korea · Department of Computer Science and Engineering Center for Brain-Machine Interface Incheon National University Incheon, South Korea

Mar 23, 2026J/K move · Enter open · S save

GeoFusion-CAD: Structure-Aware Diffusion with Geometric State Space for Parametric 3D Design

Xiaolei Zhou, Chuangjie Fang, Jie Wu, Jingyi Yang+2

Zhejiang University of Technology · Hangzhou International Innovation Institute, Beihang University

Parametric CAD Modeling 3D Diffusion Models

Open paper AskPDF arXiv

Parametric Computer-Aided Design (CAD) is fundamental to modern 3D modeling, yet existing methods struggle to generate long command sequences, especially under complex geometric and topological dependencies. Transformer-based architectures dominate CAD sequence generation due to their strong dependency modeling, but their quadratic attention cost and limited context windowing hinder scalability to long programs. We propose GeoFusion-CAD, an end-to-end diffusion framework for scalable and structure-aware generation. Our proposal encodes CAD programs as hierarchical trees, jointly capturing geometry and topology within a state-space diffusion process. Specifically, a lightweight C-Mamba block models long-range structural dependencies through selective state transitions, enabling coherent generation across extended command sequences. To support long-sequence evaluation, we introduce DeepCAD-240, an extended benchmark that increases the sequence length ranging from 40 to 240 while preserving sketch-extrusion semantics from the ABC dataset. Extensive experiments demonstrate that GeoFusion-CAD achieves superior performance on both short and long command ranges, maintaining high geometric fidelity and topological consistency where Transformer-based models degrade. Our approach sets new state-of-the-art scores for long-sequence parametric CAD generation, establishing a scalable foundation for next-generation CAD modeling systems. Code and datasets are available at GitHub.

Abstract

Parametric CAD Modeling

Explore similar work

GeoFusion-CAD: Structure-Aware Diffusion with Geometric State Space for Parametric 3D Design

DreamCAD: Scaling Multi-modal CAD Generation using Differentiable Parametric Surfaces

GuideCAD: A Lightweight Multimodal Framework for 3D CAD Model Generation via Prefix Embedding