Source Code Summarization

28 papers with code • 10 benchmarks • 7 datasets

Code Summarization is a task that tries to comprehend code and automatically generate descriptions directly from the source code.

Source: Improving Automatic Source Code Summarization via Deep Reinforcement Learning

Most implemented papers

A Transformer-based Approach for Source Code Summarization

wasiahmad/NeuralCodeSum ACL 2020

Generating a readable summary that describes the functionality of a program is known as source code summarization.

code2seq: Generating Sequences from Structured Representations of Code

tech-srl/code2seq ICLR 2019

The ability to generate natural language sequences from source code snippets has a variety of applications such as code summarization, documentation, and retrieval.

Recommendations for Datasets for Source Code Summarization

transms/m2ts NAACL 2019

The main use for these descriptions is in software documentation e. g. the one-sentence Java method descriptions in JavaDocs.

Structured Neural Summarization

CoderPat/structured-neural-summarization ICLR 2019

Summarization of long sequences into a concise statement is a core problem in natural language processing, requiring non-trivial understanding of the input.

Improving Automatic Source Code Summarization via Deep Reinforcement Learning

mf1832146/tree_transformer_2.0 17 Nov 2018

To the best of our knowledge, most state-of-the-art approaches follow an encoder-decoder framework which encodes the code into a hidden space and then decode it into natural language space, suffering from two major drawbacks: a) Their encoders only consider the sequential content of code, ignoring the tree structure which is also critical for the task of code summarization, b) Their decoders are typically trained to predict the next word by maximizing the likelihood of next ground-truth word with previous ground-truth word given.

Code Generation as a Dual Task of Code Summarization

Bolin0215/CSCGDual NeurIPS 2019

Code summarization (CS) and code generation (CG) are two crucial tasks in the field of automatic software development.

Improved Code Summarization via a Graph Neural Network

acleclair/ICPC2020_GNN 6 Apr 2020

The first approaches to use structural information flattened the AST into a sequence.

Language-Agnostic Representation Learning of Source Code from Structure and Context

danielzuegner/code-transformer ICLR 2021

Source code (Context) and its parsed abstract syntax tree (AST; Structure) are two complementary representations of the same computer program.