Combinatorial Optimization for All: Using LLMs to Aid Non-Experts in Improving Optimization Algorithms

Camilo Chacón Sartori, Christian Blum
3/14/2025
cs.AIcs.CLcs.LG

No official code found. Best implementation: camilochs/comb-opt-for-all (6 stars, last commit 2026-03-13).

Abstract

Large Language Models (LLMs) have shown notable potential in code generation for optimization algorithms, unlocking exciting new opportunities. This paper examines how LLMs, rather than creating algorithms from scratch, can improve existing ones without the need for specialized expertise. To explore this potential, we selected 10 baseline optimization algorithms from various domains (metaheuristics, reinforcement learning, deterministic, and exact methods) to solve the classic Travelling Salesman Problem. The results show that our simple methodology often results in LLM-generated algorithm variants that improve over the baseline algorithms in terms of solution quality, reduction in computational time, and simplification of code complexity, all without requiring specialized optimization knowledge or advanced algorithmic implementation skills.

View on arXivView PDF

Code Implementations(1)

camilochs/comb-opt-for-allCommunity59%

Combinatorial Optimization for All: Using LLMs to Aid Non-Experts in Improving Optimization Algorithms

61CSS, HTMLMar 13, 20257 months ago

Similar Papers

Improving Existing Optimization Algorithms with LLMs

89% match2025

Algorithm Evolution Using Large Language Model

88% match1 repo632023

DRAGON: LLM-Driven Decomposition and Reconstruction Agents for Large-Scale Combinatorial Optimization

88% match2026

Hard Constraints Meet Soft Generation: Guaranteed Feasibility for LLM-based Combinatorial Optimization

87% match2026

Can Large Language Models Invent Algorithms to Improve Themselves?: Algorithm Discovery for Recursive Self-Improvement through Reinforcement Learning

87% match2024

LLMize: A Framework for Large Language Model-Based Numerical Optimization

87% match1 repo52025

Cite this paper

@article{sartori2025combinatorial,
  title  = {Combinatorial Optimization for All: Using LLMs to Aid Non-Experts in Improving Optimization Algorithms},
  author = {Camilo Chacón Sartori and Christian Blum},
  year   = {2025},
  eprint = {2503.10968},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2503.10968v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2503.10968.svg)](https://codeofpaper.com/paper/2503.10968)

Discussion