Research map / Vision-language and multimodal / Robot action and navigation
CAD-Assistant: Tool-Augmented VLLMs as Generic CAD Task Solvers
Dimitrios Mallis, Ahmet Serda Karadeniz, Sebastian Cavada, Danila Rukhovich, Niki Foteinopoulou, Kseniya Cherenkova, Anis Kacem, Djamila Aouada
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 |
|---|---|---|---|
| use JSON point-based serialization instead of the SGPBench implicit serialized graph (ablation) | GPT-4o CAD sketch QA · SGPBench 2D CQA | accuracy: 0.674 -> 0.748 | Improved |
| use schema-embedded JSON instead of tabular HTML (ablation) | GPT-4o CAD sketch QA · SGPBench 2D CQA | accuracy: 0.710 -> 0.748 | Improved |
| use point-based instead of implicit parametrization with serialized graph (ablation) | GPT-4o CAD sketch QA · SGPBench 2D CQA | accuracy: 0.674 -> 0.744 | Improved |
| over-parametrize sketches instead of point-based only (ablation) | JSON serialization · SGPBench 2D CQA | accuracy: 0.748 -> 0.747 | No clear effect |
| give precise rendering instead of hand-drawn sketch image (ablation) | GPT-4o CAD sketch QA · SGPBench 2D CQA | accuracy: 0.616 -> 0.754 | Improved |
| use rendering instead of text-based serialization (ablation) | GPT-4o CAD sketch QA · SGPBench 2D CQA | accuracy: 0.748 -> 0.754 | Improved |
| add CAD-Assistant tool augmentation to GPT-4o planner | GPT-4o · SGPBench 2D CQA | accuracy: 0.686 -> 0.791 | Improved |
| add CAD-Assistant tool augmentation to GPT-4o planner | GPT-4o · SGPBench 3D CQA | accuracy: 0.782 -> 0.857 | Improved |
| add CAD-Assistant tool augmentation to smaller GPT-4 mini planner | GPT-4 mini · SGPBench 2D CQA | accuracy: 0.594 -> 0.614 | No clear effect |
| add multimodal sketch recognizer (MMrecog) (ablation) | CAD-Assistant 0-shot · SketchGraphs autoconstraining | CF1: 0.318 -> 0.329 | Improved |
| add constraint checker (ConstrCheck) on top of MMrecog (ablation) | CAD-Assistant 0-shot · SketchGraphs autoconstraining | CF1: 0.329 -> 0.484 | Improved |
| add 5-shot demonstrations (ablation) | CAD-Assistant · SketchGraphs autoconstraining | CF1: 0.484 -> 0.529 | Improved |
Closest papers by meaning
- CAD-Editor: A Locate-then-Infill Framework with Automated Training Data Synthesis for Text-Based CAD Editing (ICML 2025 · similarity 0.87)
- Generating CAD Code with Vision-Language Models for 3D Designs (ICLR 2025 · similarity 0.87)
- CAD-GPT: Synthesising CAD Construction Sequence with Spatial Reasoning-Enhanced Multimodal LLMs (AAAI 2025 · similarity 0.87)
- CAD-Coder: Text-to-CAD Generation with Chain-of-Thought and Geometric Reward (NeurIPS 2025 · similarity 0.87)
- Clarify Before You Draw: Proactive Agents for Robust Text-to-CAD Generation (ICML 2026 · similarity 0.86)
- GeoCAD: Local Geometry-Controllable CAD Generation with Large Language Models (NeurIPS 2025 · similarity 0.86)
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.
