SimpleEvol: An Agent-Loop Framework for LLM-Driven Automated Heuristic Design with Minimal Human Priors

2026-09-29

AI Summary

This study addresses the excessive reliance on human priors and the low intelligence conversion efficiency in the automated heuristic design of large language models (LLMs) by proposing SimpleEvol, an agent-based iterative framework. By discarding complex pipeline paradigms and manual prior interventions, this approach enables LLMs to autonomously generate and refine heuristic algorithms through iterative loops. The authors innovatively define the metrics of prior degree and intelligence conversion efficiency, systematically validating the superiority of a minimalist model-centric strategy across combinatorial optimization benchmarks. Experimental results demonstrate that SimpleEvol achieves the highest intelligence conversion efficiency on multiple tasks, significantly outperforming conventional high-prior frameworks. This work establishes an efficient new paradigm for LLM-driven automated algorithm design.

Abstract

Large language models (LLMs) have emerged as powerful tools for automated heuristic design (AHD), enabling iterative generation and refinement of heuristics. However, the dominant paradigm embeds LLMs as narrow, fixed components, such as crossover or mutation, within heavily hand-engineered evolutionary frameworks. We argue this misapprehends LLMs. It treats them as specialized tools rather than general reasoners, constrains them to low-level operations, and underutilizes their autonomy. Moreover, the extensive human priors in these frameworks violate the bitter lesson principle that general methods scaling with computation surpass hand-crafted solutions. This raises a key question: which AHD framework designs best convert stronger LLM capabilities into better heuristics? To address this, we propose metrics for LLM-driven AHD framework handcraftedness (AHI) and intelligence conversion efficiency (ICE). Evaluating ten LLMs across three challenging combinatorial optimization problems, we obtain a notable finding that frameworks with fewer human priors consistently yield higher ICE. Based on this finding, we propose SimpleEvol, an agent-loop framework for AHD which removes nearly all human priors and allows the LLM to operate autonomously. SimpleEvol consistently achieves the highest ICE, often by a large margin. Our results challenge the trend toward complex AHD pipelines and point to a lighter and more model-centric alternative, suggesting that reducing human priors is a more effective strategy to scale up with model intelligence. The source code is available at https://github.com/HenryZhu1029/SimpleEvol-Master.

Problem

• Automated Heuristic Design
• Large Language Models
• Human Priors
• Intelligence Conversion Efficiency
• Combinatorial Optimization

Innovation

• Automated Heuristic Design
• Agent-Loop Framework
• Intelligence Conversion Efficiency
• Large Language Models
• Combinatorial Optimization