Research map / Large language models / Agents, code and math
Communicating Natural Programs to Humans and Machines
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 CNN encoder for input-output conditioning (ablation) | IO+NL+pseudo-annotation synthesis · LARC synthesis test split | tasks solved: 22/183 -> 22/183 | No clear effect |
| Replace color words in descriptions with COLOR (ablation) | CLIP description pairing · LARC description pairing | pairing accuracy: 64% -> 56% | Hurt |
| Use a neural sequence decoder instead of the bigram generator (ablation) | Program synthesis on LARC · LARC synthesis test split | tasks solved | Hurt |
| Add pseudo-annotated language-program training pairs (ablation) | LARC program synthesis · LARC synthesis test split | tasks solved: 22/183 | Helped |
| Provide input-output examples as well as language instructions | Program synthesis and GPT-4 prompting on LARC · LARC | tasks solved | Helped |
Closest papers by meaning
- Think Visually, Reason Textually: Vision-Language Synergy in Abstract Reasoning (CVPR 2026 · similarity 0.87)
- Distinguishing Performance From Competence in Evaluations of Humanlike Abstract Reasoning (NeurIPS 2026 · similarity 0.86)
- Implicit Intelligence - Evaluating Agents on What Users Don’t Say (ICML 2026 · similarity 0.86)
- Product of Experts with LLMs: Boosting Performance on ARC Is a Matter of Perspective (ICML 2025 · similarity 0.86)
- On the self-verification limitations of large language models on reasoning and planning tasks (ICLR 2025 · similarity 0.86)
- RefactorBench: Evaluating Stateful Reasoning in Language Agents Through Code (ICLR 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.