Phases
3 phases
| Order | Name | Description | Activities | Actions |
|---|---|---|---|---|
| 1 | Data Preparation | Prepare a richly annotated program graph dataset ready for model training — from raw repositories through compiled projects, graph construction, … | ||
| 2 | Model Training | Learn variable semantics by running Gated Graph Neural Networks over program graphs using task-specific graph modifications and maximum likelihood training. | ||
| 3 | Evaluation | Validate model performance against baselines and ablations, and demonstrate practical relevance through real-world bug discovery. |