A Hybrid GA LLM Framework for Structured Task Optimization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shum, William, Chan, Rachel, Lin, Jonas, Feng, Benny, Lau, Patrick
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916794787168256
author Shum, William
Chan, Rachel
Lin, Jonas
Feng, Benny
Lau, Patrick
author_facet Shum, William
Chan, Rachel
Lin, Jonas
Feng, Benny
Lau, Patrick
contents GA LLM is a hybrid framework that combines Genetic Algorithms with Large Language Models to handle structured generation tasks under strict constraints. Each output, such as a plan or report, is treated as a gene, and evolutionary operations like selection, crossover, and mutation are guided by the language model to iteratively improve solutions. The language model provides domain knowledge and creative variation, while the genetic algorithm ensures structural integrity and global optimization. GA LLM has proven effective in tasks such as itinerary planning, academic outlining, and business reporting, consistently producing well structured and requirement satisfying results. Its modular design also makes it easy to adapt to new tasks. Compared to using a language model alone, GA LLM achieves better constraint satisfaction and higher quality solutions by combining the strengths of both components.
format Preprint
id arxiv_https___arxiv_org_abs_2506_07483
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Hybrid GA LLM Framework for Structured Task Optimization
Shum, William
Chan, Rachel
Lin, Jonas
Feng, Benny
Lau, Patrick
Computation and Language
GA LLM is a hybrid framework that combines Genetic Algorithms with Large Language Models to handle structured generation tasks under strict constraints. Each output, such as a plan or report, is treated as a gene, and evolutionary operations like selection, crossover, and mutation are guided by the language model to iteratively improve solutions. The language model provides domain knowledge and creative variation, while the genetic algorithm ensures structural integrity and global optimization. GA LLM has proven effective in tasks such as itinerary planning, academic outlining, and business reporting, consistently producing well structured and requirement satisfying results. Its modular design also makes it easy to adapt to new tasks. Compared to using a language model alone, GA LLM achieves better constraint satisfaction and higher quality solutions by combining the strengths of both components.
title A Hybrid GA LLM Framework for Structured Task Optimization
topic Computation and Language
url https://arxiv.org/abs/2506.07483