NatGen: generative pre-training by "naturalizing" source code.
Saikat ChakrabortyToufique AhmedYangruibo DingPremkumar T. DevanbuBaishakhi RayPublished in: ESEC/SIGSOFT FSE (2022)
Keyphrases
- source code
- software systems
- open source
- software projects
- software maintenance
- open source software
- static analysis
- mining software repositories
- object oriented systems
- plagiarism detection
- program comprehension
- software evolution
- version control
- open source projects
- change impact analysis
- software repositories
- software artifacts
- maintenance activities
- execution traces
- impact analysis
- software engineers
- artificial intelligence
- manual inspection
- text files
- open source software projects
- linux kernel
- authorship attribution
- bug reports
- code examples
- legacy systems
- visual basic
- high level