PLLM: Pseudo-Labeling Large Language Models for CAD Program Synthesis
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914327776198656 |
|---|---|
| author | Li, Yuanbo Shu, Dule Chen, Yanying Klenk, Matt Ritchie, Daniel |
| author_facet | Li, Yuanbo Shu, Dule Chen, Yanying Klenk, Matt Ritchie, Daniel |
| contents | Recovering Computer-Aided Design (CAD) programs from 3D geometries is a widely studied problem. Recent advances in large language models (LLMs) have enabled progress in CAD program synthesis, but existing methods rely on supervised training with paired shape-program data, which is often unavailable. We introduce PLLM, a self-training framework for CAD program synthesis from unlabeled 3D shapes. Given a pre-trained CAD-capable LLM and a shape dataset, PLLM iteratively samples candidate programs, selects high-fidelity executions, and augments programs to construct synthetic program-shape pairs for fine-tuning. We experiment on adapting CAD-Recode from DeepCAD to the unlabeled ABC dataset show consistent improvements in geometric fidelity and program diversity. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_12561 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | PLLM: Pseudo-Labeling Large Language Models for CAD Program Synthesis Li, Yuanbo Shu, Dule Chen, Yanying Klenk, Matt Ritchie, Daniel Computer Vision and Pattern Recognition Recovering Computer-Aided Design (CAD) programs from 3D geometries is a widely studied problem. Recent advances in large language models (LLMs) have enabled progress in CAD program synthesis, but existing methods rely on supervised training with paired shape-program data, which is often unavailable. We introduce PLLM, a self-training framework for CAD program synthesis from unlabeled 3D shapes. Given a pre-trained CAD-capable LLM and a shape dataset, PLLM iteratively samples candidate programs, selects high-fidelity executions, and augments programs to construct synthetic program-shape pairs for fine-tuning. We experiment on adapting CAD-Recode from DeepCAD to the unlabeled ABC dataset show consistent improvements in geometric fidelity and program diversity. |
| title | PLLM: Pseudo-Labeling Large Language Models for CAD Program Synthesis |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2602.12561 |