Research map / Graph learning / Graph algorithms and clustering
TreeCaps: Tree-Based Capsule Networks for Source Code Processing
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 |
|---|---|---|---|
| Use VTS routing instead of DRSW routing (ablation) | TreeCaps · SA dataset (SrcML, type+token) | accuracy: 84.57 -> 85.43 | Helped |
| Use VTS routing instead of DRSW routing (ablation) | TreeCaps · OJ dataset (PycParser, type+token) | accuracy: 96.74 -> 98.32 | Helped |
| Use VTS routing instead of DRSW routing (ablation) | TreeCaps · OJ dataset (SrcML, type+token) | accuracy: 87.77 -> 88.40 | Helped |
| Use VTS routing instead of DRSW routing (ablation) | TreeCaps · Java-Small | F1: 42.89 -> 46.78 | Helped |
| Use VTS routing instead of DRSW routing (ablation) | TreeCaps · Java-Med | F1: 52.56 -> 55.67 | Helped |
| Use VTS routing instead of DRSW routing (ablation) | TreeCaps · Java-Large | F1: 57.82 -> 61.34 | Helped |
| Use VTS routing instead of DRSW routing (ablation) | TreeCaps · OJ dataset | validation accuracy convergence | Helped |
| Use VTS routing instead of DRSW routing (ablation) | TreeCaps · Java-Large test set (average transformations) | PPC: 14.51% -> 12.49% | Helped |
| Combine node type and token information instead of using type information alone (ablation) | TreeCaps-VTS · SA dataset (SrcML) | accuracy: 84.60 -> 85.43 | Helped |
| Combine node type and token information instead of using token information alone (ablation) | TreeCaps-VTS · SA dataset (SrcML) | accuracy: 78.15 -> 85.43 | Helped |
| Combine node type and token information instead of using type information alone (ablation) | TreeCaps-VTS · OJ dataset (PycParser) | accuracy: 95.88 -> 98.32 | Helped |
| Combine node type and token information instead of using token information alone (ablation) | TreeCaps-VTS · OJ dataset (PycParser) | accuracy: 90.21 -> 98.32 | Helped |
Closest papers by meaning
- Convolutional Neural Networks over Tree Structures for Programming Language Processing (AAAI 2016 · similarity 0.88)
- Tree-to-tree Neural Networks for Program Translation (NeurIPS 2018 · similarity 0.87)
- Learning to Represent Programs with Graphs (ICLR 2018 · similarity 0.86)
- TreeGen: A Tree-Based Transformer Architecture for Code Generation (AAAI 2020 · similarity 0.86)
- Global Relational Models of Source Code (ICLR 2020 · similarity 0.85)
- Diffusion On Syntax Trees For Program Synthesis (ICLR 2025 · similarity 0.83)
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.