CodeChain: Towards Modular Code Generation Through Chain of Self-revisions with Representative Sub-modules

arXiv:2310.08992v3 Announce Type: replace-cross
Abstract: Large Language Models (LLMs) have already become quite proficient at solving simpler programming tasks like those in HumanEval or MBPP benchmarks. However, solving more complex and competitive programming tasks is still quite challenging for these models - possibly due to their tendency to generate solutions as monolithic code blocks instead of decomposing them into logical sub-tasks and sub-modules. On the other hand, experienced programmers instinctively write modularized code with abstraction for solving complex tasks, often …

abstract arxiv become benchmarks code code generation cs.ai cs.cl cs.pl however humaneval language language models large language large language models llms modular modules programming tasks through type