Inferring Javascript types using Graph Neural Networks

2019/05/16 by Jessica Schrouff, Schrouff, Jessica, Kai Wohlfahrt +5 · 1 voice · 2 citations
Computer Science · #Advanced Malware Detection Techniques #Software Engineering Research #Software Testing and Debugging Techniques #cs.LG #cs.PL #cs.SE #stat.ML

paper · pdf · doi:10.48550/arxiv.1905.06707

openalex publication_date 2019/05/16 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

The recent use of `Big Code' with state-of-the-art deep learning methods offers promising avenues to ease program source code writing and correction. As a first step towards automatic code repair, we implemented a graph neural network model that predicts token types for Javascript programs. The predictions achieve an accuracy above 90%, which improves on previous similar work.

Citations

• Graph Convolutional Matrix Completion

Cited by

• Adversarial Robustness for Code
• Learning to Execute Programs with Instruction Pointer Attention Graph\n Neural Networks

Discussions

• Inferring JavaScript Types Using Graph Neural Networks [hn, 4 points, 0 comments]

Related

• Call Me Maybe: Enhancing JavaScript Call Graph Construction using Graph Neural Networks
• Advanced Graph-Based Deep Learning for Probabilistic Type Inference
• Breaking Obfuscation: Cluster-Aware Graph with LLM-Aided Recovery for Malicious JavaScript Detection
• The Hidden DNA of LLM-Generated JavaScript: Structural Patterns Enable High-Accuracy Authorship Attribution
• Learning Semantic Program Embeddings with Graph Interval Neural Network