ASE 2024
Sun 27 October - Fri 1 November 2024 Sacramento, California, United States
Thu 31 Oct 2024 14:15 - 14:30 at Carr - Code generation 3 Chair(s): Jialun Cao

Although large language models (LLMs) have demonstrated impressive ability in code generation, they are still struggling to address the complicated intent provided by humans. It is widely acknowledged that humans typically employ planning to decompose complex problems and schedule solution steps prior to implementation. To this end, we introduce planning into code generation to help the model understand complex intent and reduce the difficulty of problem-solving. This paper proposes a self-planning code generation approach with large language models, which consists of two phases, namely planning phase and implementation phase. Specifically, in the planning phase, LLM plans out concise solution steps from the intent combined with few-shot prompting. Subsequently, in the implementation phase, the model generates code step by step, guided by the preceding solution steps. We conduct extensive experiments on various code-generation benchmarks across multiple programming languages. Experimental results show that self-planning code generation achieves a relative improvement of up to 25.4% in Pass@1 compared to direct code generation, and up to 11.9% compared to Chain-of-Thought of code generation. Moreover, our self-planning approach also enhances the quality of the generated code with respect to correctness, readability, and robustness, as assessed by humans.

Thu 31 Oct

Displayed time zone: Pacific Time (US & Canada) change

13:30 - 15:00
Code generation 3Industry Showcase / Journal-first Papers / Research Papers at Carr
Chair(s): Jialun Cao Hong Kong University of Science and Technology
13:30
15m
Talk
Test-Driven Development and LLM-based Code Generation
Research Papers
Noble Saji Mathews University of Waterloo, Canada, Mei Nagappan University of Waterloo
13:45
15m
Talk
A Pair Programming Framework for Code Generation via Multi-Plan Exploration and Feedback-Driven RefinementACM SigSoft Distinguished Paper Award
Research Papers
Huan Zhang Nanjing University, Wei Cheng Nanjing University, Yuhan Wu Nanjing University, Wei Hu Nanjing University
14:00
15m
Talk
Ansible Lightspeed: A Code Generation Service for IT Automation
Industry Showcase
Priyam Sahoo , Saurabh Pujar IBM Research AI, Ganesh Nalawade RED HAT, Richard Gebhardt , Louis Mandel IBM Research, USA, Luca Buratti IBM Research
Link to publication DOI Pre-print File Attached
14:15
15m
Talk
Self-planning Code Generation with Large Language Models
Journal-first Papers
Xue Jiang , Yihong Dong Peking University, Lecheng Wang Peking University, Fang Zheng Peking University, Qiwei Shang Peking University, Ge Li Peking University, Zhi Jin Peking University, Wenpin Jiao Peking University