PanGu-Coder2: Boosting Large Language Models for Code with Ranking Feedback

• DOI: 10.48550/arxiv.2307.14936
• Publication Date: 2023-01-01
• Authors: Shen, Bo; Zhang, Jiaxin; Chen, Taihong; Zan, Daoguang; Geng, Bing; Fu, An; Zeng, Muhan; Yu, Ailun; Ji, Jichuan; Zhao, Jingyang; Guo, Yuenan; Wang, Qianxiang
• Keywords: Software Engineering (cs.SE); FOS: Computer and information sciences; Computer Science - Machine Learning; Computer Science - Software Engineering; Computer Science - Computation and Language; Computer Science - Programming Languages; Artificial Intelligence (cs.AI); Computer Science - Artificial Intelligence; Computation and Language (cs.CL); Machine Learning (cs.LG); Programming Languages (cs.PL)

Preprint<br/>Large Language Models for Code (Code LLM) are flourishing. New and powerful models are released on a weekly basis, demonstrating remarkable performance on the code generation task. Various approaches have been proposed to boost the code generation performance of pre-trained Code LLMs, such as supervised fine-tuning, instruction tuning, reinforcement learning, etc. In this paper, we propose a novel RRTF (Rank Responses to align Test&Teacher Feedback) framework, which can effectively and efficiently boost pre-trained large language models for code generation. Under this framework, we present PanGu-Coder2, which achieves 62.20% pass@1 on the OpenAI HumanEval benchmark. Furthermore, through an extensive evaluation on CoderEval and LeetCode benchmarks, we show that PanGu-Coder2 consistently outperforms all previous Code LLMs.<br/>