Research map / Architectures and efficiency / Neural architecture search
EvoPrompting: Language Models for Code-Level Neural Architecture Search
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 |
|---|---|---|---|
| search with EvoPrompting to discover new triplet representations instead of the standard Triplet-GMPNN | Triplet-GMPNN · CLRS Heapsort OOD | OOD accuracy: 31.0 -> 69.9 | Helped |
| search with EvoPrompting to discover new triplet representations | Triplet-GMPNN · CLRS Bubble Sort OOD | OOD accuracy: 67.7 -> 88.9 | Helped |
| search with EvoPrompting to discover new triplet representations | Triplet-GMPNN · CLRS BFS OOD | OOD accuracy: 99.7 -> 100.0 | No clear effect |
| remove prompt-tuning between rounds (ablation) | EvoPrompting · MNIST-1D | max fitness / Pareto frontier | Hurt |
| select parents randomly instead of by fitness (ablation) | EvoPrompting · MNIST-1D | max fitness | Hurt |
| add prompt-tuning to random-parents prompting (ablation) | naive few-shot prompting · MNIST-1D | fitness | No clear effect |
| evolutionary search with prompt-tuning over naive few-shot prompting (ablation) | PaLM 62B · MNIST-1D | Pareto frontier of test error vs model size | Helped |
Closest papers by meaning
- EvoOptiGraph: Weakness-Driven Coevolution via Graph-Based Structural Generation for Optimization Modeling (NeurIPS 2026 · similarity 0.88)
- How to Auto-optimize Prompts for Domain Tasks? Adaptive Prompting and Reasoning through Evolutionary Domain Knowledge Adaptation (NeurIPS 2025 · similarity 0.87)
- Simple Baselines are Competitive with Code Evolution (NeurIPS 2026 · similarity 0.86)
- Use Your INSTINCT: INSTruction optimization for LLMs usIng Neural bandits Coupled with Transformers (ICML 2024 · similarity 0.86)
- ReEvo: Large Language Models as Hyper-Heuristics with Reflective Evolution (NeurIPS 2024 · similarity 0.86)
- Scaling Prompt Synthesis for Large Language Model Reasoning (ICML 2026 · 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.