Research map / Large language models / Agents, code and math
Learning to Infer Program Sketches
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 |
|---|---|---|---|
| Combine a neural sketch generator with symbolic synthesis | AlgoLisp · AlgoLisp full dataset Test | accuracy: 88.6 -> 90.0 | Helped |
| Combine a neural sketch generator with symbolic synthesis | AlgoLisp · AlgoLisp filtered Test | accuracy: 95.6 -> 95.8 | No clear effect |
| Use natural-language descriptions rather than input-output examples as the learned specification (ablation) | SketchAdapt · AlgoLisp full dataset Test | accuracy: 8.3 -> 90.0 | Helped |
| Use natural-language descriptions rather than input-output examples as the learned specification (ablation) | SketchAdapt · AlgoLisp filtered Test | accuracy: 8.8 -> 95.8 | Helped |
| Combine learned pattern recognition with symbolic search | AlgoLisp unseen Even programs · AlgoLisp generalization: Even | accuracy: 4.5 -> 34.4 | Helped |
| Combine learned pattern recognition with symbolic search | AlgoLisp unseen Odd programs · AlgoLisp generalization: Odd | accuracy: 1.1 -> 29.8 | Helped |
| Use SketchAdapt rather than Generator only | String transformations with noisy specifications · String transformations with noisy specification | accuracy: 52% -> 53% | No clear effect |
| Replace natural-language descriptions with input-output examples (ablation) | SketchAdapt · AlgoLisp | accuracy: 90.0 -> 8.3 | Hurt |
Closest papers by meaning
- PLANS: Neuro-Symbolic Program Learning from Videos (NeurIPS 2020 · similarity 0.90)
- Neural Sketch Learning for Conditional Program Generation (ICLR 2018 · similarity 0.89)
- Representing Partial Programs with Blended Abstract Semantics (ICLR 2021 · similarity 0.88)
- Leveraging Language to Learn Program Abstractions and Search Heuristics (ICML 2021 · similarity 0.88)
- Neural-Guided Deductive Search for Real-Time Program Synthesis from Examples (ICLR 2018 · similarity 0.88)
- Program Synthesis via Test-Time Transduction (NeurIPS 2025 · similarity 0.86)
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.