BlenderLLM: Training Large Language Models for Computer-Aided Design with Self-improvement
This work addresses the underexplored area of LLMs in CAD, providing a foundation for CAD automation, though it appears incremental as it builds on existing self-improvement methodologies.
The paper tackles the problem of applying Large Language Models (LLMs) to Computer-Aided Design (CAD) tasks, where existing models have significant limitations in generating accurate CAD scripts, and shows that BlenderLLM, using minimal instruction-based fine-tuning and iterative self-improvement, significantly surpasses these models in functionality and accuracy.
The application of Large Language Models (LLMs) in Computer-Aided Design (CAD) remains an underexplored area, despite their remarkable advancements in other domains. In this paper, we present BlenderLLM, a novel framework for training LLMs specifically for CAD tasks leveraging a self-improvement methodology. To support this, we developed a bespoke training dataset, BlendNet, and introduced a comprehensive evaluation suite, CADBench. Our results reveal that existing models demonstrate significant limitations in generating accurate CAD scripts. However, through minimal instruction-based fine-tuning and iterative self-improvement, BlenderLLM significantly surpasses these models in both functionality and accuracy of CAD script generation. This research establishes a strong foundation for the application of LLMs in CAD while demonstrating the transformative potential of self-improving models in advancing CAD automation. We encourage further exploration and adoption of these methodologies to drive innovation in the field. The dataset, model, benchmark, and source code are publicly available at https://github.com/FreedomIntelligence/BlenderLLM