Design-Specification Tiling for ICL-based CAD Code Generation
This addresses a domain-specific problem for CAD code generation, offering an incremental improvement over existing exemplar selection methods.
The paper tackles the problem of poor performance of large language models in generating Computer-Aided Design (CAD) code due to scarce training data by proposing Design-Specification Tiling, a method for selecting exemplars in In-Context Learning that improves CAD code generation quality, consistently outperforming existing strategies.
Large language models (LLMs) have demonstrated remarkable capabilities in code generation, yet they underperform on domain-specific tasks such as Computer-Aided Design (CAD) code generation due to scarce training data. In-Context Learning (ICL) offers a training-free alternative through task-specific exemplars. However, existing selection strategies prioritize similarity or point-wise diversity, often producing redundant selections that fail to satisfy the compositional requirements of complex CAD design specifications. In this work, we propose knowledge sufficiency as a principled objective for exemplar selection that aims to maximally satisfy all requirements within design specifications. To realize this objective, we introduce Design-Specification Tiling (DST), which quantifies knowledge sufficiency through a surrogate tiling ratio by extracting multi-granular design components and measuring the proportion of query components covered by selected exemplars. We demonstrate that maximizing this objective constitutes submodular maximization and provide a polynomial-time greedy algorithm with a (1-1/e)-approximation guarantee. Extensive experiments demonstrate that DST substantially improves CAD code generation quality, consistently outperforming existing exemplar selection strategies in ICL.