Summary of Self-explained Keywords Empower Large Language Models For Code Generation, by Lishui Fan and Mouxiang Chen and Zhongxin Liu
Self-Explained Keywords Empower Large Language Models for Code Generation
by Lishui Fan, Mouxiang Chen, Zhongxin Liu
First submitted to arxiv on: 21 Oct 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Software Engineering (cs.SE)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed technique, SEK (Self-Explained Keywords), aims to improve code generation in large language models (LLMs) by extracting and explaining key terms in problem descriptions. This is achieved by ranking these terms based on frequency, allowing LLMs to focus on high-frequency keywords rather than low-frequency ones. The authors conducted comprehensive experiments across three benchmarks using five representative LLMs, demonstrating significant and consistent gains in code generation accuracy. For example, SEK improved the Pass@1 of DeepSeek-Coder-V2-Instruct from 85.4% to 93.3% on the HumanEval benchmark. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Large language models can generate impressive codes, but they often struggle with low-frequency keywords. To help them, researchers developed a new technique called SEK (Self-Explained Keywords). This method helps LLMs understand what’s important in a problem description and focus on high-frequency words instead of low-frequency ones. The results show that this technique can greatly improve the accuracy of generated code. |
