2020/03/31 by Vivek Gupta, Vivek Kumar Gupta, Gupta, Vivek · 4 citations
Computer Science · #Abstract syntax tree #Artificial intelligence #Artificial neural network #Code (set theory) #Computation and Language (cs.CL) #Computer science #Documentation #FOS: Computer and information sciences #Heuristics #Machine Learning (cs.LG) #Machine learning #Programming language #Set (abstract data type) #Software Engineering Research #Software engineering #Source code #Syntax #Topic Modeling #Transformer #Web Data Mining and Analysis #cs.CL #cs.LG
paper · pdf · doi:10.48550/arxiv.2004.00998
published in arXiv (Cornell University) (Cornell University) · arXiv admin note: substantial text overlap with arXiv:1902.01954 by other authors
arxiv created 2020/03/31 · openalex publication_date 2020/03/31 · arxiv updated 2020/04/03 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Source code summarizing is a task of writing short, natural language descriptions of source code behavior during run time. Such summaries are extremely useful for software development and maintenance but are expensive to manually author,hence it is done for small fraction of the code that is produced and is often ignored. Automatic code documentation can possibly solve this at a low cost. This is thus an emerging research field with further applications to program comprehension, and software maintenance. Traditional methods often relied on cognitive models that were built in the form of templates and by heuristics and had varying degree of adoption by the developer community. But with recent advancements, end to end data-driven approaches based on neural techniques have largely overtaken the traditional techniques. Much of the current landscape employs neural translation based architectures with recurrence and attention which is resource and time intensive training procedure. In this paper, we employ neural techniques to solve the task of source code summarizing and specifically compare NMT based techniques to more simplified and appealing Transformer architecture on a dataset of Java methods and comments. We bring forth an argument to dispense the need of recurrence in the training procedure. To the best of our knowledge, transformer based models have not been used for the task before. With supervised samples of more than 2.1m comments and code, we reduce the training time by more than 50% and achieve the BLEU score of 17.99 for the test set of examples.