code2seq: Generating Sequences from Structured Representations of Code

TL;DR

code2seq introduces a novel method that leverages the syntactic structure of code, using AST paths and attention mechanisms, to improve sequence generation tasks like code summarization and documentation.

Contribution

It presents a new approach that encodes source code as AST paths, outperforming previous models in code sequence generation tasks across multiple datasets.

Findings

Significantly outperforms previous models on code summarization tasks.

Effective across two programming languages and four datasets.

Utilizes syntactic structure for better code representation.

Abstract

The ability to generate natural language sequences from source code snippets has a variety of applications such as code summarization, documentation, and retrieval. Sequence-to-sequence (seq2seq) models, adopted from neural machine translation (NMT), have achieved state-of-the-art performance on these tasks by treating source code as a sequence of tokens. We present ${\rm {\scriptsize CODE2SEQ}}$: an alternative approach that leverages the syntactic structure of programming languages to better encode source code. Our model represents a code snippet as the set of compositional paths in its abstract syntax tree (AST) and uses attention to select the relevant paths while decoding. We demonstrate the effectiveness of our approach for two tasks, two programming languages, and four datasets of up to $16$M examples. Our model significantly outperforms previous models that were specifically designed for programming languages, as well as state-of-the-art NMT models. An interactive online demo of our model is available at http://code2seq.org. Our code, data and trained models are available at http://github.com/tech-srl/code2seq.