EvoCAD: Evolutionary CAD Code Generation with Vision Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Preintner, Tobias, Yuan, Weixuan, König, Adrian, Bäck, Thomas, Raponi, Elena, van Stein, Niki
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914090585161728
author Preintner, Tobias
Yuan, Weixuan
König, Adrian
Bäck, Thomas
Raponi, Elena
van Stein, Niki
author_facet Preintner, Tobias
Yuan, Weixuan
König, Adrian
Bäck, Thomas
Raponi, Elena
van Stein, Niki
contents Combining large language models with evolutionary computation algorithms represents a promising research direction leveraging the remarkable generative and in-context learning capabilities of LLMs with the strengths of evolutionary algorithms. In this work, we present EvoCAD, a method for generating computer-aided design (CAD) objects through their symbolic representations using vision language models and evolutionary optimization. Our method samples multiple CAD objects, which are then optimized using an evolutionary approach with vision language and reasoning language models. We assess our method using GPT-4V and GPT-4o, evaluating it on the CADPrompt benchmark dataset and comparing it to prior methods. Additionally, we introduce two new metrics based on topological properties defined by the Euler characteristic, which capture a form of semantic similarity between 3D objects. Our results demonstrate that EvoCAD outperforms previous approaches on multiple metrics, particularly in generating topologically correct objects, which can be efficiently evaluated using our two novel metrics that complement existing spatial metrics.
format Preprint
id arxiv_https___arxiv_org_abs_2510_11631
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle EvoCAD: Evolutionary CAD Code Generation with Vision Language Models
Preintner, Tobias
Yuan, Weixuan
König, Adrian
Bäck, Thomas
Raponi, Elena
van Stein, Niki
Computer Vision and Pattern Recognition
Artificial Intelligence
Neural and Evolutionary Computing
Combining large language models with evolutionary computation algorithms represents a promising research direction leveraging the remarkable generative and in-context learning capabilities of LLMs with the strengths of evolutionary algorithms. In this work, we present EvoCAD, a method for generating computer-aided design (CAD) objects through their symbolic representations using vision language models and evolutionary optimization. Our method samples multiple CAD objects, which are then optimized using an evolutionary approach with vision language and reasoning language models. We assess our method using GPT-4V and GPT-4o, evaluating it on the CADPrompt benchmark dataset and comparing it to prior methods. Additionally, we introduce two new metrics based on topological properties defined by the Euler characteristic, which capture a form of semantic similarity between 3D objects. Our results demonstrate that EvoCAD outperforms previous approaches on multiple metrics, particularly in generating topologically correct objects, which can be efficiently evaluated using our two novel metrics that complement existing spatial metrics.
title EvoCAD: Evolutionary CAD Code Generation with Vision Language Models
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Neural and Evolutionary Computing
url https://arxiv.org/abs/2510.11631