REMoH: A Reflective Evolution of Multi-objective Heuristics approach via Large Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Forniés-Tabuenca, Diego, Uribe, Alejandro, Otamendi, Urtzi, Artetxe, Arkaitz, Rivera, Juan Carlos, de Lacalle, Oier Lopez
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913885472161792
author Forniés-Tabuenca, Diego
Uribe, Alejandro
Otamendi, Urtzi
Artetxe, Arkaitz
Rivera, Juan Carlos
de Lacalle, Oier Lopez
author_facet Forniés-Tabuenca, Diego
Uribe, Alejandro
Otamendi, Urtzi
Artetxe, Arkaitz
Rivera, Juan Carlos
de Lacalle, Oier Lopez
contents Multi-objective optimization is fundamental in complex decision-making tasks. Traditional algorithms, while effective, often demand extensive problem-specific modeling and struggle to adapt to nonlinear structures. Recent advances in Large Language Models (LLMs) offer enhanced explainability, adaptability, and reasoning. This work proposes Reflective Evolution of Multi-objective Heuristics (REMoH), a novel framework integrating NSGA-II with LLM-based heuristic generation. A key innovation is a reflection mechanism that uses clustering and search-space reflection to guide the creation of diverse, high-quality heuristics, improving convergence and maintaining solution diversity. The approach is evaluated on the Flexible Job Shop Scheduling Problem (FJSSP) in-depth benchmarking against state-of-the-art methods using three instance datasets: Dauzere, Barnes, and Brandimarte. Results demonstrate that REMoH achieves competitive results compared to state-of-the-art approaches with reduced modeling effort and enhanced adaptability. These findings underscore the potential of LLMs to augment traditional optimization, offering greater flexibility, interpretability, and robustness in multi-objective scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2506_07759
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle REMoH: A Reflective Evolution of Multi-objective Heuristics approach via Large Language Models
Forniés-Tabuenca, Diego
Uribe, Alejandro
Otamendi, Urtzi
Artetxe, Arkaitz
Rivera, Juan Carlos
de Lacalle, Oier Lopez
Artificial Intelligence
Neural and Evolutionary Computing
I.2.7; I.2.8; F.2.2
Multi-objective optimization is fundamental in complex decision-making tasks. Traditional algorithms, while effective, often demand extensive problem-specific modeling and struggle to adapt to nonlinear structures. Recent advances in Large Language Models (LLMs) offer enhanced explainability, adaptability, and reasoning. This work proposes Reflective Evolution of Multi-objective Heuristics (REMoH), a novel framework integrating NSGA-II with LLM-based heuristic generation. A key innovation is a reflection mechanism that uses clustering and search-space reflection to guide the creation of diverse, high-quality heuristics, improving convergence and maintaining solution diversity. The approach is evaluated on the Flexible Job Shop Scheduling Problem (FJSSP) in-depth benchmarking against state-of-the-art methods using three instance datasets: Dauzere, Barnes, and Brandimarte. Results demonstrate that REMoH achieves competitive results compared to state-of-the-art approaches with reduced modeling effort and enhanced adaptability. These findings underscore the potential of LLMs to augment traditional optimization, offering greater flexibility, interpretability, and robustness in multi-objective scenarios.
title REMoH: A Reflective Evolution of Multi-objective Heuristics approach via Large Language Models
topic Artificial Intelligence
Neural and Evolutionary Computing
I.2.7; I.2.8; F.2.2
url https://arxiv.org/abs/2506.07759