Research map / Large language models / Agents, code and math
Sharing State Between Prompts and Programs
Ellie Y Cheng, Logan Weber, Tian Jin, Michael Carbin
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 specialization to Python and optimizations (eager variable loading, caching) to Nightjar (ablation) | Nightjar (Baseline) · SPSBench | runtime overhead: reduces runtime overhead by 2.0-2.1x | Improved |
| use shared program state instead of manual implementation | Nightjar · SPSBench | lines of code: 42.0 -> 25.1 | Improved |
| use shared program state instead of manual implementation | Nightjar · SPSBench | task accuracy: +4-19% | Improved |
| use shared program state instead of manual implementation | Nightjar · SPSBench | runtime: 0.4-4.3x manual implementations | Hurt |
Closest papers by meaning
- BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructions (ICLR 2025 · similarity 0.86)
- When Prompt Engineering Meets Software Engineering: CNL-P as Natural and Robust "APIs'' for Human-AI Interaction (ICLR 2025 · similarity 0.84)
- Beyond Accuracy: Evaluating Self-Consistency of Code Large Language Models with IdentityChain (ICLR 2024 · similarity 0.83)
- Large Language Models as Tool Makers (ICLR 2024 · similarity 0.83)
- CRAFT: Customizing LLMs by Creating and Retrieving from Specialized Toolsets (ICLR 2024 · similarity 0.83)
- SemCoder: Training Code Language Models with Comprehensive Semantics Reasoning (NeurIPS 2024 · 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.
