EUVIMEDEUVIMED entdecken
Crossref

ReCAD: Reinforcement Learning Enhanced Parametric CAD Model Generation with Vision-Language Models

Jiahao Li, Yusheng Luo, Yunzhong Lou, Xiangdong Zhou

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (8) · 6190-6198

DOI: 10.1609/aaai.v40i8.37544 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Jiahao Li und 3 weiteren Autor:innen. Die Arbeit wurde bisher 8-mal zitiert (Crossref).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We present ReCAD, a reinforcement learning (RL) framework that bootstraps pretrained large models (PLMs) to generate precise parametric computer-aided design (CAD) models from multimodal inputs by leveraging their inherent generative capabilities. With just access to simple functional interfaces (e.g., point coordinates), our approach enables the emergence of complex CAD operations (e.g., pattern replication and mirror). This stands in contrast to previous methods, which typically rely on knowledge injected through supervised fine-tuning (SFT), offer limited support for editability, and fail to exploit the strong generative priors of PLMs. Specifically, the ReCAD framework begins by fine-tuning vision-language models (VLMs) to equip them with basic CAD model generation capabilities, where we rewrite CAD scripts into parameterized code that is leveraged to generate accurate textual descriptions for supervision. Then, we propose a novel RL strategy that incorporates parameterized code as guidance to enhance the model’s reasoning on challenging questions. Furthermore, we employ a hierarchical primitive learning process to progressively teach structured and compositional skills under a unified reward function that ensures both geometric accuracy and semantic fidelity. ReCAD sets a new state-of-the-art in both text-to-CAD and image-to-CAD tasks, significantly improving geometric accuracy across in-distribution and out-of-distribution settings. In the image-to-CAD task, for instance, it reduces the mean Chamfer Distance from 73.47 to 29.61 (in-distribution) and from 272.06 to 80.23 (out-of-distribution), outperforming existing baselines by a substantial margin.

Zitieren

APA

Li, J., Luo, Y., Lou, Y., & Zhou, X. (2026). ReCAD: Reinforcement Learning Enhanced Parametric CAD Model Generation with Vision-Language Models. Proceedings of the AAAI Conference on Artificial Intelligence, 40(8), 6190-6198. https://doi.org/10.1609/aaai.v40i8.37544

Vancouver

Li J, Luo Y, Lou Y, Zhou X. ReCAD: Reinforcement Learning Enhanced Parametric CAD Model Generation with Vision-Language Models. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(8):6190-6198. doi:10.1609/aaai.v40i8.37544

BibTeX
@article{Li2026,
  author = {Li, Jiahao and Luo, Yusheng and Lou, Yunzhong and Zhou, Xiangdong},
  title = {ReCAD: Reinforcement Learning Enhanced Parametric CAD Model Generation with Vision-Language Models},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {8},
  pages = {6190--6198},
  doi = {10.1609/aaai.v40i8.37544}
}
Kostenlos starten →