Awesome AI4CAD Hub Papers · Datasets · Projects
← Back to papers

Clarify Before You Draw: Proactive Agents for Robust Text-to-CAD Generation

arXiv 2026 67.4 method

TLDR

ProCAD proactively clarifies ambiguous text prompts before generating parametric CAD programs, improving robustness and reducing errors.

Reasoning

The paper introduces a novel proactive agentic framework that addresses ambiguity in text-to-CAD generation, showing strong empirical gains. However, it focuses narrowly on CadQuery and does not evaluate on other CAD representations or tasks like reconstruction.

Read-first score

Read-first score 67.4, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 78.

Recency 8%
100

Uses a gentle age decay so recent papers surface without erasing older foundations. 2026

Reproducibility 25%
85

Screens links and visible text for paper, code, dataset, artifact, and repository signals. pdf=True; code=True; dataset=False; markers=code,dataset,github

Methodology quality 25%
70

Screens visible abstract and analysis fields for experiment, dataset, baseline, metric, and limitation evidence. markers=dataset,experiment,metric

Topical relevance 42%
48.8

Uses existing LLM keyword relevance scores normalized to 0-100. AI for CAD,computer-aided design,neural CAD,generative CAD,parametric CAD,B-Rep,boundary representation,constructive solid geometry,CSG,sketch extrusion,CAD generation,CAD reconstruction,text-to-CAD,image-to-CAD,point cloud to CAD,CAD program

Field roles

FrontierMethodology anchorReproducibility anchor

Rank sensitivity

Stability: volatile; rank range: 17.

Keyword Scores

text-to-CAD
10
parametric CAD
9
CAD generation
9
CAD program
9
AI for CAD
8
generative CAD
8
computer-aided design
7
neural CAD
5
sketch extrusion
3
constructive solid geometry
2
CSG
2
CAD reconstruction
2
B-Rep
1
boundary representation
1
image-to-CAD
1
point cloud to CAD
1

Deep Analysis

Innovations

  • Proactive clarifying agent that audits prompts and asks targeted clarification questions only when necessary
  • Agentic framework pairing a clarifying agent with a CAD coding agent to resolve specification issues before code synthesis
  • Fine-tuning the coding agent on a curated high-quality text-to-CadQuery dataset and training the clarifying agent via agentic SFT on clarification trajectories

Methodology

ProCAD pairs a proactive clarifying agent that audits the prompt and asks targeted clarification questions to produce a self-consistent specification, with a CAD coding agent that translates the specification into an executable CadQuery program. The coding agent is fine-tuned on a curated text-to-CadQuery dataset, and the clarifying agent is trained via agentic supervised fine-tuning on clarification trajectories.

Key Results

Proactive clarification significantly improves robustness to ambiguous prompts with low interaction overhead; ProCAD reduces mean Chamfer distance by 79.9% and lowers invalidity ratio from 4.8% to 0.9% compared to Claude Sonnet 4.5.

Tags

LG