Design-Specification Tiling for ICL-based CAD Code Generation

2026-03-13

AI Summary

Large language models face significant limitations in specialized domains such as CAD code generation due to scarce training data, and existing in-context learning (ICL) approaches struggle to select examples that adequately satisfy the compositional requirements of complex design specifications. This work introduces “knowledge sufficiency” as a novel optimization objective for ICL example selection and proposes the Design Specification Tiling (DST) method. DST extracts design components at multiple granularities, quantifies the coverage of candidate examples relative to the query’s requirements via a tiling ratio, and formulates the selection problem as a submodular maximization task, for which a theoretically grounded greedy algorithm is developed. Experimental results demonstrate that DST substantially outperforms existing strategies, significantly improving both the quality and specification compliance of the generated code.

Technology Category

• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

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.

Problem

• CAD code generation
• In-Context Learning
• exemplar selection
• design specification
• compositional requirements

Innovation

• Design-Specification Tiling
• In-Context Learning
• Knowledge Sufficiency
• Submodular Maximization
• CAD Code Generation