Generative Code Modeling with Graphs
This work addresses the challenge of code generation for developers and AI systems, presenting an incremental improvement over existing methods.
The paper tackles the problem of generating source code by modeling it as a structured prediction task, introducing a model that uses graphs to represent intermediate states and interleaves grammar-driven expansion with neural message passing, resulting in outperforming strong baselines in generating semantically meaningful expressions.
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.