LLMs show promise in generating genetic programming operators

Researchers have evaluated the ability of eight large language models (LLMs) to generate effective parent-selection operators for genetic programming (GP) in symbolic regression tasks. The study found that models like Claude Sonnet 4.6 and Gemini 3.1 Pro consistently performed well, with one operator synthesized by Kimi K2.5 outperforming standard automatic selection methods. These findings suggest that LLMs can be used in a zero-shot manner to create competitive GP selection operators, often by leveraging semantic understanding to guide the search process. AI

IMPACT Demonstrates LLMs' potential for zero-shot synthesis of specialized operators in computational tasks like genetic programming.

RANK_REASON Academic paper detailing a novel application of LLMs to a specific computational task. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.NE (Neural & Evolutionary) →

• arXiv
• Claude Sonnet 4.6
• Gemini 3.1 Pro
• Genetic Programming
• Kimi K2.5
• Large Language Models
• Symbolic Regression

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →