Generative Code Modeling with Graphs
arXiv:1805.08490
Abstract
Generative models for source code are an interesting structured prediction problem, requiring to reason about both hard syntactic and semantic constraints as well as about natural, likely programs. We present a novel model for this problem that uses a graph to represent the intermediate state of the generated output. The generative procedure interleaves grammar-driven expansion steps with graph augmentation and neural message passing steps. An experimental evaluation shows that our new model can generate semantically meaningful expressions, outperforming a range of strong baselines.
References in corpus (5)
Cited by in corpus (17)
- Natural Attack for Pre-trained Models of Code
- GraphCodeBERT: Pre-training Code Representations with Data Flow
- Graph Neural Networks Meet Neural-Symbolic Computing: A Survey and Perspective
- CODIT: Code Editing with Tree-Based Neural Models
- Neural Reverse Engineering of Stripped Binaries using Augmented Control Flow Graphs
- Detecting Code Clones with Graph Neural Networkand Flow-Augmented Abstract Syntax Tree
- Embedding API Dependency Graph for Neural Code Generation
- Neural Bug Finding: A Study of Opportunities and Challenges
- CS1QA: A Dataset for Assisting Code-based Question Answering in an Introductory Programming Course
- Learning Autocompletion from Real-World Datasets
- Representing Partial Programs with Blended Abstract Semantics
- World Programs for Model-Based Learning and Planning in Compositional State and Action Spaces
- How could Neural Networks understand Programs?
- Using GGNN to recommend log statement level
- Learning to Extend Program Graphs to Work-in-Progress Code
- Unsupervised Program Synthesis for Images By Sampling Without Replacement
- Hashing Modulo Alpha-Equivalence