cs.AINov 17, 2026

GRAFT: Execution-Guided Genetic Recombination for Agent Design

Authors: Sevda Ebrahimi, Mohammed Saidul Islam, Xiaodan Zhu, Elham Dolatabadi

Organizations: York University · Vector Institute · Queen’s University

Abstract

Automated agent design searches for executable architectures that improve LLM-based agent performance, with evolutionary methods using language-model crossover to recombine promising mechanisms for task decomposition, model coordination, and verification. This search, however, is typically guided by aggregate fitness alone, which hides the instance-level successes and failures behind each score. We introduce Execution Guided Genetic Recombination for Agent Design (GRAFT), which uses aggregate fitness for selection and survival and parent-specific execution evidence to guide semantic crossover with integrated mutation. Its feedback harness reuses bounded samples of successful and failed runs from fitness evaluation, withholding reference answers and requiring no additional task executions. We evaluate GRAFT across six reasoning and programming settings using open-weight optimizer and executor models. It achieves a mean test accuracy of 82.0%, exceeding the strongest baseline by 9.8% points. Controlled ablations show that adding execution evidence improves test accuracy with initialization, search rules, repair policies, and evaluation budgets held fixed. Our results support the value of parent-specific execution evidence for evolutionary agent design.

Figures & tables

Appendix figures & tables1 asset

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. VeriEvolve: Evaluator-First Evolution for Tool-Using Agents

    Oct 21, 2026Yueshen Li, Kamer Ali Yuksel, Gokhan Tur +2Self-Evolving AgentsTool-Use Evaluation

  2. EvoAgentBench: Benchmarking Agent Self-Evolution via Ability Transfer

    Jul 6, 2026Xingze Gao, Chuanrui Hu, Hongda Chen +9LLM Agent Self-ImprovementAI Agent Benchmarks