Computer Science
a year ago

New code generation framework reduces errors and boosts code reuse rates!

arXiv (Cornell University)
Context-Aware Code Generation Framework for Code Repositories: Local, Global, and Third-Party Library Awareness
Dianshu Liao, Shidong Pan, Qing Huang, Xiaoxue Ren, Zhenchang Xing, Huan Jin, Qinying Li

Paper Summary

Related papers

Theme background image
Computer Science
a year ago

New code generation framework reduces errors and boosts code reuse rates!

Empirical Studies
Log Analysis
Scientific Workflows
1 view
Theme background image
Computer Science
2 years ago

Revolutionizing Code Merging: A Game-Changer for Model-Driven Development!

Model-Driven Engineering
Empirical Studies
Autonomic Computing
2 views
Theme background image
Computer Science
3 years ago

Revolutionizing Software Development: Code Repository Enhances Efficiency and Innovation

Empirical Studies
Software Testing
Software Reliability
1 view

Code generation tools are important for software developers, but current tools often don't consider the context of the code repository, leading to errors and compatibility issues. The A^3-CodGen framework uses information from the code repository to generate more accurate and efficient code. By incorporating local, global, and third-party library information, the framework produces code with fewer errors and higher reusability. This approach results in code that is more reliable and reusable compared to code written by human developers.