ChatCoder: Chat-based Refine Requirement Improves LLMs' Code Generation
This addresses the challenge for developers and users in effectively communicating requirements to AI systems for code generation, though it is incremental as it builds on existing refine-based and fine-tuning methods.
The paper tackles the problem of vague, incomplete, and ambiguous human requirements in natural language that cause large language models to make mistakes in code generation, and proposes ChatCoder, a chat-based method to refine these requirements, resulting in a large margin improvement in performance over existing models.
Large language models have shown good performances in generating code to meet human requirements. However, human requirements expressed in natural languages can be vague, incomplete, and ambiguous, leading large language models to misunderstand human requirements and make mistakes. Worse, it is difficult for a human user to refine the requirement. To help human users refine their requirements and improve large language models' code generation performances, we propose ChatCoder: a method to refine the requirements via chatting with large language models. We design a chat scheme in which the large language models will guide the human users to refine their expression of requirements to be more precise, unambiguous, and complete than before. Experiments show that ChatCoder has improved existing large language models' performance by a large margin. Besides, ChatCoder has the advantage over refine-based methods and LLMs fine-tuned via human response.