.png){kind=icon}
返回
LLM-Based Test-Driven Interactive Code Generation: User Study and Empirical Evaluation
.png){kind=icon}
OA
AI
DOI：10.1109/TSE.2024.3428972.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Large language models (LLMs) have shown great potential in automating significant aspects of coding by producing natural code from informal natural language (NL) intent. However, given NL is informal, it does not lend easily to checking that the generated code correctly satisfies the user intent. In this paper, we propose a novel interactive workflow TiCoder for guided intent clarification (i.e., partial formalization) through tests to support the generation of more accurate code suggestions. Through a mixed methods user study with 15 programmers, we present an empirical evaluation of the effectiveness of the workflow to improve code generation accuracy. We find that participants using the proposed workflow are significantly more likely to correctly evaluate AI generated code, and report significantly less task-induced cognitive load. Furthermore, we test the potential of the workflow at scale with four different state-of-the-art LLMs on two python datasets, using an idealized proxy for a user feedback. We observe an average absolute improvement of 45.97% in the pass@1 code generation accuracy for both datasets and across all LLMs within 5 user interactions, in addition to the automatic generation of accompanying unit tests.
Keyword:
Intent disambiguation
code generation
LLMs
human factors
cognitive load
test generation
Intent disambiguation
code generation
LLMs
human factors
cognitive load
test generation
期刊
IF:
5.6
论文数:
2.9K
被引数:
1.1W
机构
引用论文
.png){kind=icon}
分享收藏Autophagic Gene Polymorphisms in Liquid Biopsies and Outcome of Patients with Metastatic Clear Cell Renal Cell Carcinoma自噬基因多态性在转移性透明细胞肾细胞癌患者的液体活检和预后中的作用
OAAI
OAAI分享收藏
查看更多内容