A Hybrid GA LLM Framework for Structured Task Optimization

TL;DR

This paper introduces GA LLM, a hybrid framework combining Genetic Algorithms and Large Language Models to optimize structured tasks with constraints, improving solution quality and adaptability.

Contribution

It presents a novel hybrid approach that leverages the strengths of GAs and LLMs for structured task optimization, with demonstrated effectiveness across multiple domains.

Findings

Outperforms standalone LLMs in constraint satisfaction

Produces higher quality, well-structured outputs

Adapts easily to new structured tasks

Abstract

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.