Comparing PushGP and GPT-4o on Program Synthesis with only Input-Output Examples

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

@InProceedings{hernandez:2025:GECCOcomp2,

• author = "Jose Guadalupe Hernandez and Anil Kumar Saini and Gabriel Ketron and Jason Moore",
• title = "Comparing {PushGP} and {GPT-4o} on Program Synthesis with only Input-Output Examples",
• booktitle = "Proceedings of the 2025 Genetic and Evolutionary Computation Conference Companion",
• year = "2025",
• editor = "Aniko Ekart and Nelishia Pillay",
• pages = "619--622",
• 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, program synthesis, large language models, programming by examples: Poster",
• isbn13 = "979-8-4007-1464-1",
• URL = "https://doi.org/10.1145/3712255.3726700",
• DOI = "10.1145/3712255.3726700",
• size = "4 pages",
• abstract = "Genetic programming (GP) and large language models (LLMs) have both achieved notable success in program synthesis. However, the methods for specifying the desired program behavior (i.e., user intent) differ: GP relies on input-output examples, whereas LLMs use text descriptions. In this work, we compare the capabilities of a GP system, PushGP, and an LLM model, GPT-4o, in synthesizing programs where the user intent is specified through input-output examples. Using tasks from the PSB2 program synthesis benchmark, we found that PushGP solved more tasks than GPT-4o. While some tasks were successfully solved by both synthesizers, others were uniquely solved by only one of them, highlighting their complementary strengths. In addition to the prompt with just input-output examples (data-only), we tested GPT-4o with another prompt containing only a textual description of the task (text-only). Both prompt variants successfully solved the same 7 tasks (with different success rates), with the data-only prompt solving an additional task. Ultimately, each synthesizer is successful in distinct ways, highlighting differences in their underlying methodologies.",
• notes = "GECCO-2025 GP A Recombination of the 34th International Conference on Genetic Algorithms (ICGA) and the 30th Annual Genetic Programming Conference (GP)",

}

Genetic Programming entries for Jose Guadalupe Hernandez Anil Kumar Saini Gabriel Ketron Jason H Moore

Citations