LEAR: LLM-Driven Evolution of Agent-Based Rules

Created by W.Langdon from gp-bibliography.bib Revision:1.8620

@InProceedings{gurkan:2025:GECCOcomp,

• author = "Can Gurkan and Narasimha Karthik Jwalapuram and Kevin Wang and Rudy Danda and Leif Rasmussen and John Chen and Uri Wilensky",
• title = "{LEAR:} {LLM-Driven} Evolution of Agent-Based Rules",
• booktitle = "Large Language Models for and with Evolutionary Computation Workshop",
• year = "2025",
• editor = "Erik Hemberg and Roman Senkerik and Joel Lehman and Una-May O'Reilly and Michal Pluhacek and Niki {van Stein} and Pier Luca and Tome Eftimov",
• pages = "2309--2326",
• address = "Malaga, Spain",
• series = "GECCO '25 Companion",
• month = "14-18 " # jul,
• organisation = "SIGEVO",
• publisher = "Association for Computing Machinery",
• publisher_address = "New York, NY, USA",
• keywords = "genetic algorithms, genetic programming, large language models, evolutionary computation, multi-agent systems, agent-based modeling, ANN",
• isbn13 = "979-8-4007-1464-1",
• URL = "https://doi.org/10.1145/3712255.3734368",
• DOI = "10.1145/3712255.3734368",
• size = "18 pages",
• abstract = "This study investigates the feasibility and effectiveness of integrating Large Language Models (LLMs) as mutation operators within Genetic Programming (GP) frameworks so as to evolve agent behaviors in multi-agent systems (MAS) and provide benchmarks that evaluate the efficacy of LLM-generated code in multi-agent domains. Our approach leverages the sophisticated code-generation capabilities of LLMs to introduce semantically meaningful variations during the evolutionary process. Specifically, we explore and systematically compare these different prompting strategies: zero-shot, one-shot, and two-shot prompting as well as prompting the generation of commented code to assess their impact on the quality of evolved agent behaviors. Additionally, we propose a novel methodology where evolution operates at a higher abstraction level by mutating pseudocode representations of agent behaviors, subsequently converting them into executable code through another LLM-mediated step. This strategy capitalizes on the extensive natural language training data of LLMs, potentially enabling the discovery of more innovative solutions. Our results indicate that LLM-driven mutation with comment generation enhances agent performance while mutating pseudocode representations yields reduced performance. This research contributes valuable insights regarding the integration of LLM-driven GP techniques into MAS, highlighting both the potential and limitations of these approaches. All code is open-sourced at https://github.com/can-gurkan/LEAR.",
• notes = "GECCO-2025 LLMfwEC workshop A Recombination of the 34th International Conference on Genetic Algorithms (ICGA) and the 30th Annual Genetic Programming Conference (GP)",

}

Genetic Programming entries for Can Gurkan Narasimha Karthik Jwalapuram Kevin Wang Rudy Danda Leif Rasmussen John Chen Uri Wilensky

Citations