4 ms·
Learning to Represent Edits
- telotortium 7y ago> ABSTRACT > > We introduce the problem of learning distributed representations of edits. By combining a “neural editor” with an “edit encoder”, our models learn to represent the salient information of an edit and can be used to apply edits to new inputs. We experiment on natural language and source code edit data. Our evaluation yields promising results that suggest that our neural network models learn to capture the structure and semantics of edits. We hope that this interesting task and data source will inspire other researchers to work further on this problem.
- TheAceOfHearts 7y agoSomeone should add [PDF] to the title. It would also provide better context by including "within the context of AI / ML" (not sure which term would be most appropriate) at the end of the title. Under the Tasks section they talk about building assistive coding tools. The given example is to consume the huge GitHub commit stream in order to identify patterns for certain kinds of bug fixes and "best practices". You'd always want to have a human review all changes, but this sounds like it could be used to help you migrate and modernize old code.
- tobylane 7y agoBetter link https://openreview.net/forum?id=BJl6AjC5F7 https://openreview.net/forum?id=BJl6AjC5F7