HyEvo: Self-Evolving Hybrid Agentic Workflows for Efficient Reasoning

ArXi:2603.19639v1 Announce Type: new Although agentic workflows have nstrated strong potential for solving complex tasks, existing automated generation methods remain inefficient and underperform, as they rely on predefined operator libraries and homogeneous LLM-only workflows in which all task-level computation is performed through probabilistic inference. To address these limitations, we propose HyEvo, an automated workflow-generation framework that leverages heterogeneous atomic synthesis.