Research map / Large language models / Agents, code and math
Automated Data Extraction Using Predictive Program Synthesis
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 operator-specific lifting functions to bottom-up synthesis (ablation) | ColumnSplit · text extraction, 20 benchmarks | execution time (s): 210.73 -> 4.20 | Helped |
| add operator-specific lifting functions for filter operators (ablation) | WebExtract · web extraction, 20 webpages | execution time (s): 27.04 -> 6.41 | Helped |
| use full set of data-type regex tokens instead of five basic tokens (ablation) | ColumnSplit · text extraction, 20 benchmarks | test cases with all fields extracted: 14 of 20 -> full token set | Helped |
| allow iterative splitting of extracted columns | ColumnSplit · text extraction, 20 benchmarks | fields extracted | Helped |
Closest papers by meaning
- Disjunctive Program Synthesis: A Robust Approach to Programming by Example (AAAI 2018 · similarity 0.86)
- Synthesis of Programs from Multimodal Datasets (AAAI 2018 · similarity 0.85)
- Generating Pragmatic Examples to Train Neural Program Synthesizers (ICLR 2024 · similarity 0.83)
- HYSYNTH: Context-Free LLM Approximation for Guiding Program Synthesis (NeurIPS 2024 · similarity 0.83)
- ExeDec: Execution Decomposition for Compositional Generalization in Neural Program Synthesis (ICLR 2024 · similarity 0.83)
- Program Synthesis via Test-Time Transduction (NeurIPS 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.