TabPFN for Zero-shot Parametric Engineering Design Generation

2026-02-02

AI Summary

This work addresses the high computational cost, heavy reliance on large training datasets, and limited adaptability of conventional engineering design generation methods to changing requirements or data. We propose a novel zero-shot, training-free conditional generation framework by introducing TabPFN into parametric engineering design. The approach requires only a few reference samples to generate designs conditioned on target performance metrics, without any task-specific fine-tuning. Evaluated on hull, Blended-Wing-Body aircraft, and UIUC airfoil datasets, the method achieves design performance errors below 2%, while demonstrating strong diversity and robustness. Moreover, its computational overhead is substantially lower than that of diffusion models, significantly reducing dependence on both data volume and computational resources.

Technology Category

• Natural Language Processing: Generation
• Machine Learning: Deep Generative Models & Autoencoders
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Deep generative models for engineering design often require substantial computational cost, large training datasets, and extensive retraining when design requirements or datasets change, limiting their applicability in real-world engineering design workflow. In this work, we propose a zero-shot generation framework for parametric engineering design based on TabPFN, enabling conditional design generation using only a limited number of reference samples and without any task-specific model training or fine-tuning. The proposed method generates design parameters sequentially conditioned on target performance indicators, providing a flexible alternative to conventional generative models. The effectiveness of the proposed approach is evaluated on three engineering design datasets, i.e., ship hull design, BlendedNet aircraft, and UIUC airfoil. Experimental results demonstrate that the proposed method achieves competitive diversity across highly structured parametric design spaces, remains robust to variations in sampling, resolution and parameter dimensionality of geometry generation, and achieves a low performance error (e.g., less than 2% in generated ship hull designs'performance). Compared with diffusion-based generative models, the proposed framework significantly reduces computational overhead and data requirements while preserving reliable generation performance. These results highlight the potential of zero-shot, data-efficient generation as a practical and efficient tool for engineering design, enabling rapid deployment, flexible adaptation to new design settings, and ease of integration into real-world engineering workflows.

Problem

• engineering design
• generative models
• computational cost
• data efficiency
• zero-shot generation

Innovation

• TabPFN
• zero-shot generation
• parametric design
• data-efficient generative modeling
• engineering design automation