Research map / Graph learning / Knowledge and temporal graphs
Open Vocabulary Learning on Source Code with a Graph-Structured Cache
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add a graph-structured cache with word-use edges | AugAST-GGNN · Fill-In-The-Blank, Seen repos | accuracy: 0.90 -> 0.97 | Helped |
| Add a graph-structured cache with word-use edges | AugAST-GGNN · Fill-In-The-Blank, Unseen repos | accuracy: 0.84 -> 0.92 | Helped |
| Add a graph-structured cache with word-use edges | AugAST-GGNN · Variable Naming, Seen repos | accuracy: 0.26 -> 0.53 | Helped |
| Add a graph-structured cache with word-use edges | AugAST-GGNN · Variable Naming, Unseen repos | accuracy: 0.08 -> 0.41 | Helped |
| Add data- and control-flow semantic edges (ablation) | GSC code model · Fill-In-The-Blank, Seen repos | accuracy: 0.89 -> 0.97 | Helped |
| Add data- and control-flow semantic edges (ablation) | GSC code model · Fill-In-The-Blank, Unseen repos | accuracy: 0.80 -> 0.92 | Helped |
| Add data- and control-flow semantic edges (ablation) | GSC code model · Variable Naming, Seen repos | accuracy: 0.49 -> 0.53 | Helped |
| Add data- and control-flow semantic edges (ablation) | GSC code model · Variable Naming, Unseen repos | accuracy: 0.38 -> 0.41 | Helped |
| Use GSC instead of a closed vocabulary (ablation) | AugAST-GGNN · Variable Naming, Unseen repos | accuracy: 0.04 -> 0.41 | Helped |
| Use GSC instead of CharCNN-only word handling (ablation) | AugAST-GGNN · Variable Naming, Unseen repos | accuracy: 0.06 -> 0.41 | Helped |
| Use cache-node connections to word uses, rather than pointer output without those edges (ablation) | AugAST variable-naming model · Variable Naming, Seen repos | accuracy: 0.26 -> 0.53 | Helped |
| Use cache-node connections to word uses, rather than pointer output without those edges (ablation) | AugAST variable-naming model · Variable Naming, Unseen repos | accuracy: 0.08 -> 0.41 | Helped |
Closest papers by meaning
- Learning to Represent Programs with Graphs (ICLR 2018 · similarity 0.88)
- GraphCodeBERT: Pre-training Code Representations with Data Flow (ICLR 2021 · similarity 0.86)
- PLUR: A Unifying, Graph-Based View of Program Learning, Understanding, and Repair (NeurIPS 2021 · similarity 0.86)
- Can Machines Read Coding Manuals Yet? – A Benchmark for Building Better Language Models for Code Understanding (AAAI 2022 · similarity 0.86)
- LILO: Learning Interpretable Libraries by Compressing and Documenting Code (ICLR 2024 · similarity 0.85)
- Structured Packing in LLM Training Improves Long Context Utilization (AAAI 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.