PulseAugur
LIVE 08:16:02
research · [3 sources] ·
0
research

AI agents generate dynamic CAD models and million-scale programs

Researchers have developed new agentic systems for Computer-Aided Design (CAD) that can generate complex 3D assemblies with moving parts, a capability previously lacking in AI-driven design tools. One system, AADvark, incorporates external solvers and visual feedback to reason about dynamic interactions, enabling the creation of functional mechanical components. Concurrently, the Zero-to-CAD framework synthesizes approximately one million executable CAD sequences, providing a large-scale dataset of procedural design information. This synthetic data has been used to train vision-language models that can reconstruct editable CAD programs from images, outperforming existing baselines. AI

Summary written by gemini-2.5-flash-lite from 3 sources. How we write summaries →

IMPACT Advances in agentic CAD systems could accelerate the design and iteration of complex mechanical components and provide valuable synthetic datasets for training future AI models.

RANK_REASON The cluster contains two arXiv papers detailing new research and frameworks for AI-assisted CAD model generation.

Read on arXiv cs.CV →

COVERAGE [3]

  1. arXiv cs.AI TIER_1 · Mitch Adler, Matthew Russo, Michael Cafarella ·

    Agent-Aided Design for Dynamic CAD Models

    arXiv:2604.15184v2 Announce Type: replace Abstract: In the past year, researchers have created agentic systems that can design real-world CAD-style objects in a training-free setting, a new variety of system that we call Agent-Aided Design. These systems place an agent in a feedb…

  2. arXiv cs.CV TIER_1 · Mohammadmehdi Ataei, Farzaneh Askari, Kamal Rahimi Malekshan, Pradeep Kumar Jayaraman ·

    Zero-to-CAD: Agentic Synthesis of Interpretable CAD Programs at Million-Scale Without Real Data

    arXiv:2604.24479v1 Announce Type: new Abstract: Computer-Aided Design (CAD) models are defined by their construction history: a parametric recipe that encodes design intent. However, existing large-scale 3D datasets predominantly consist of boundary representations (B-Reps) or me…

  3. arXiv cs.CV TIER_1 · Pradeep Kumar Jayaraman ·

    Zero-to-CAD: Agentic Synthesis of Interpretable CAD Programs at Million-Scale Without Real Data

    Computer-Aided Design (CAD) models are defined by their construction history: a parametric recipe that encodes design intent. However, existing large-scale 3D datasets predominantly consist of boundary representations (B-Reps) or meshes, stripping away this critical procedural in…