CADEvolve：透過程式演化創建真實CAD模型

摘要

電腦輔助設計（CAD）為工程與製造領域提供快速可編輯的建模方案。隨著人工智慧技術的突破，各類CAD任務現已可實現全自動化處理。然而，數據匱乏成為發展瓶頸：現有公開資料庫多數僅包含草圖拉伸序列，缺乏複雜操作、多工序組合及設計意圖表徵，導致模型微調效果受限。雖然有研究嘗試採用凍結視覺語言模型繞過此限制，但現有基礎模型對三維空間理解不足，常生成過於簡單或無效的程序。我們提出CADEvolve——基於演化演算法的流程與數據集，從簡單幾何基元出發，透過VLM引導的編輯與驗證機制，逐步建構出符合工業級複雜度的CAD程序。最終生成8,000個以可執行的CadQuery參數化生成器表達的複雜零件。經過多階段後處理與數據擴充，我們獲得包含130萬組腳本與渲染幾何配對的統一數據集，完整覆蓋CadQuery操作集。基於CADEvolve微調的VLM在Image2CAD任務中，於DeepCAD、Fusion 360和MCB三大基準測試均達到最先進水平。

English

Computer-Aided Design (CAD) delivers rapid, editable modeling for engineering and manufacturing. Recent AI progress now makes full automation feasible for various CAD tasks. However, progress is bottlenecked by data: public corpora mostly contain sketch-extrude sequences, lack complex operations, multi-operation composition and design intent, and thus hinder effective fine-tuning. Attempts to bypass this with frozen VLMs often yield simple or invalid programs due to limited 3D grounding in current foundation models. We present CADEvolve, an evolution-based pipeline and dataset that starts from simple primitives and, via VLM-guided edits and validations, incrementally grows CAD programs toward industrial-grade complexity. The result is 8k complex parts expressed as executable CadQuery parametric generators. After multi-stage post-processing and augmentation, we obtain a unified dataset of 1.3m scripts paired with rendered geometry and exercising the full CadQuery operation set. A VLM fine-tuned on CADEvolve achieves state-of-the-art results on the Image2CAD task across the DeepCAD, Fusion 360, and MCB benchmarks.

CADEvolve：透過程式演化創建真實CAD模型

CADEvolve: Creating Realistic CAD via Program Evolution

摘要

Support