SRLCG introduces a groundbreaking framework that enables large language models to generate complete, functional project code for users with minimal coding knowledge.

Key Innovations:

• Multidimensional Chain-of-Thought approach that guides LLMs through complex project development
• Dynamic Backtracking mechanism that detects and corrects errors during generation
• Autonomous generation of complete, functional projects instead of isolated code snippets
• Bridges the gap for non-technical users who lack the expertise to modify and integrate AI-generated code

Engineering Impact: This research fundamentally changes how non-technical users can leverage AI for software development, potentially democratizing programming by removing the need for extensive coding expertise to build functional applications.

SRLCG: Self-Rectified Large-Scale Code Generation with Multidimensional Chain-of-Thought and Dynamic Backtracking