Co-evolving Agent Architectures and Interpretable Reasoning for Automated Optimization

TL;DR

This paper introduces EvoOR-Agent, a co-evolutionary framework that enhances automated optimization by evolving agent architectures and reasoning paths, improving performance and interpretability over existing methods.

Contribution

The paper presents a novel co-evolutionary approach that explicitly models and evolves agent workflows and reasoning trajectories for automated optimization tasks.

Findings

EvoOR-Agent outperforms zero-shot LLMs and fixed-pipeline agents on OR benchmarks.

Explicit architecture evolution improves both performance and interpretability.

Graph-supported reasoning search contributes to better optimization results.

Abstract

Automating operations research (OR) with large language models (LLMs) remains limited by hand-crafted reasoning--execution workflows. Complex OR tasks require adaptive coordination among problem interpretation, mathematical formulation, solver selection, code generation, and iterative debugging. To address this limitation, we propose EvoOR-Agent, a co-evolutionary framework for automated optimization. The framework represents agent workflows as activity-on-edge (AOE)-style networks, making workflow topology, execution dependencies, and alternative reasoning paths explicit. On this representation, the framework maintains an architecture graph and evolves a population of reasoning individuals through graph-mediated path-conditioned recombination, multi-granularity semantic mutation, and elitist population update. A knowledge-base-assisted experience-acquisition module further injects reusable OR practices into initialization and semantic variation. Empirical results on heterogeneous OR benchmarks show that the proposed framework consistently improves over zero-shot LLMs, fixed-pipeline OR agents, and representative evolutionary agent frameworks. Case studies and ablation analyses further indicate that explicit architecture evolution and graph-supported reasoning-trajectory search contribute to both performance improvement and structural interpretability. These results suggest that treating agent architectures and reasoning trajectories as evolvable objects provides an effective route toward adaptive and interpretable automated optimization.