codingClaire / Structural-Code-Understanding

A Survey of Deep Learning Models for Structural Code Understanding

Code Understanding Literatures in Deep Learning

Check Our Survey on arxiv!

Sequence-based Models

Total: 44 papers

By years

• 2021:4 paper(s)
• 2020:12 paper(s)
• 2019:7 paper(s)
• 2018:8 paper(s)
• 2017:3 paper(s)
• 2016:6 paper(s)
• 2014:3 paper(s)
• 2012:1 paper(s)

By tasks

• Program Classification :2 paper(s)
• Code Search:3 paper(s)
• Code Generation:19 paper(s)
• Pretrain:4 paper(s)
• Code representation:1 paper(s)
• Safety Analysis:4 paper(s)
• Program Repair :2 paper(s)
• Clone Detection :2 paper(s)
• Code Summarization:7 paper(s)

By datasets

• Java:6 paper(s)
• DeepFix:1 paper(s)
• TFix's Code Patches Data:1 paper(s)
• C:6 paper(s)
• 9714 Java projects from GitHub:1 paper(s)
• Python:2 paper(s)
• JavaScript:1 paper(s)
• JS150:2 paper(s)
• python:1 paper(s)
• Uncategorized:44 paper(s)
• PY150:2 paper(s)
• C#:4 paper(s)
• 10072 Java GitHub repositories:1 paper(s)
• C#(dataset of CodeNN):0 paper(s)

By models

• N-gram:3 paper(s)
• TreeBERT:1 paper(s)
• Others:1 paper(s)
• word2vec:2 paper(s)
• Multinomial Naive Bayes (MNB) :0 paper(s)
• GRU:3 paper(s)
• Bi-LSTM:4 paper(s)
• word embedding:1 paper(s)
• Transformer:8 paper(s)
• CRF:1 paper(s)
• DNN:1 paper(s)
• pointer network:1 paper(s)
• DBN:2 paper(s)
• CAN:1 paper(s)
• LSTM:15 paper(s)
• RNN:5 paper(s)

Graph-based Models

Total: 36 papers

By years

• 2021:9 paper(s)
• 2020:10 paper(s)
• 2019:9 paper(s)
• 2018:3 paper(s)
• 2017:1 paper(s)
• 2016:1 paper(s)
• 2015:1 paper(s)
• 2014:2 paper(s)

By tasks

• Defect Prediction:3 paper(s)
• Code Search:2 paper(s)
• Program Repair:6 paper(s)
• Code Generation:5 paper(s)
• Program Verification:1 paper(s)
• Program Classification:4 paper(s)
• Vulnerability Detection:3 paper(s)
• Clone Detection:8 paper(s)
• Code Summarization:10 paper(s)

By datasets

• Java repos collected in this work:1 paper(s)
• Code-Change-Data:1 paper(s)
• Hybrid-DeepCom Dataset:1 paper(s)
• JAVA method naming datasets:1 paper(s)
• ARM binary dataset:1 paper(s)
• Genius Dataset:1 paper(s)
• Google Code Jam (GCJ):0 paper(s)
• notebookcdg:1 paper(s)
• program variables dataset produced in this work:1 paper(s)
• gcc dataset:1 paper(s)
• Python method documentation dataset:1 paper(s)
• JS150:2 paper(s)
• Findutils:1 paper(s)
• Validation dataset:1 paper(s)
• Devign Dataset:1 paper(s)
• C Dataset:1 paper(s)
• Diffutils:1 paper(s)
• OpenCL Dataset:1 paper(s)
• code-comment pairs:1 paper(s)
• BCB:1 paper(s)
• collected in this work:4 paper(s)
• Coreutils:1 paper(s)
• DeepFix dataset:1 paper(s)
• Syntax similar dataset:1 paper(s)
• SPoC:1 paper(s)
• IJDataset2.0:1 paper(s)
• CodeForces dataset:1 paper(s)
• Linux kernel's code collected in this work:1 paper(s)
• OJClone:5 paper(s)
• YANCFG Dataset:1 paper(s)
• Defects4J:1 paper(s)
• PY150:3 paper(s)
• iclr18-prog-graphs-dataset:1 paper(s)
• TL-CodeSum:1 paper(s)
• CoCoNet:1 paper(s)
• MSKCFG Dataset:1 paper(s)
• C Program Dataset:1 paper(s)
• Java method-comment pairs:1 paper(s)
• BigCloneBench:2 paper(s)
• Firmware image dataset:1 paper(s)
• C# dataset:2 paper(s)
• CodeSearchNet:2 paper(s)
• Java Dataset collected in this work:1 paper(s)
• C dataset:1 paper(s)

By models

• GINN:1 paper(s)
• Tree-RNN:1 paper(s)
• GRU:3 paper(s)
• Text-associated DeepWalk:1 paper(s)
• Multi-Relational Graph Neural Network:1 paper(s)
• TBCNN:1 paper(s)
• Flow2Vec:1 paper(s)
• LSTM:5 paper(s)
• DGCNN:1 paper(s)
• GNN:11 paper(s)
• Transformer:3 paper(s)
• Structure2vec:1 paper(s)
• MPNN:2 paper(s)
• GAT:3 paper(s)
• CNN:7 paper(s)
• GCN:2 paper(s)
• RNN:3 paper(s)
• Tree-LSTM:2 paper(s)
• tree-based LSTM:1 paper(s)
• Decision tree:1 paper(s)
• HAConvGNN:1 paper(s)
• ConvGNN:2 paper(s)
• GTN:1 paper(s)
• GGNN:8 paper(s)
• CharCNN:1 paper(s)
• Feed-forward neural network:1 paper(s)
• bidirectional RNN:1 paper(s)
• code property graphs:1 paper(s)
• attention:1 paper(s)
• Attention mechanism:1 paper(s)