You no longer have to worry about how to work in code in ACL. Our extensive solution guarantees simple and fast document management, allowing you to work on ACL documents in a few moments instead of hours or days. Our platform contains all the tools you need: merging, adding fillable fields, signing forms legally, placing shapes, and much more. There’s no need to install additional software or bother with pricey programs demanding a powerful device. With only two clicks in your browser, you can access everything you need.
Start now and manage all various types of files professionally!
Hi, my name is Wasi Ahmad and I am presenting our work, A Transformer-based Approach for Source Code Summarization. This is joint work with Saikat Chakraborty, Baishakhi Ray from Columbia University, and my advisor Kai-Wei Chang from UCLA. Source code summarization refers to the task of creating human-readable summaries that describe the functionality of a program. With the progress of natural language generation using neural sequence-to-sequence learning, recent approaches in literature frame code summarization as translating a piece of source code into a short natural language description. For example, given this Python source code snippet, a code summarization model should be able to generate a summary, similar to the human-written one. In this work, we study the Transformer, a sequence generation model that has been found effective in many natural language generation applications but hasnt been explored in source code summarization. A notable amount of prior works in source code
