Research map / Image generation / Autoregressive and latent generation
Neural Program Generation Modulo Static Analysis
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 |
|---|---|---|---|
| Condition method-body generation on static-analysis attributes | NSG versus CodeGPT · Java method generation | Pass all checks: 67.73% -> 86.41% | Helped |
| Condition method-body generation on static-analysis attributes (ablation) | NSG versus CNG · Java method generation | Pass all checks: 12.87% -> 86.41% | Helped |
| Condition method-body generation on static-analysis attributes | NSG versus GPTNeo1.3B · Java method generation | Set of API Calls: 37% -> 53% | Helped |
| Condition method-body generation on static-analysis attributes (ablation) | NSG versus CNG · Java method generation | Sequences of Program Paths: 17% -> 39% | Helped |
| Condition method-body generation on static-analysis attributes | NSG versus GPTNeo1.3B · Java method generation | AST Exact Match: 15% -> 26% | Helped |
| Condition method-body generation on static-analysis attributes | NSG versus Codex · Java method generation | No type errors: 91.42% -> 97.08% | Helped |
Closest papers by meaning
- Neural Sketch Learning for Conditional Program Generation (ICLR 2018 · similarity 0.90)
- Leveraging Grammar and Reinforcement Learning for Neural Program Synthesis (ICLR 2018 · similarity 0.86)
- Neuro-Symbolic Program Synthesis (ICLR 2017 · similarity 0.86)
- Gradient-Based Program Synthesis with Neurally Interpreted Languages (ICLR 2026 · similarity 0.86)
- Generative Code Modeling with Graphs (ICLR 2019 · similarity 0.85)
- Execution Guided Line-by-Line Code Generation (NeurIPS 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.