AI RESEARCH
FlowSteer: Towards Agents Designing Agentic Workflows via Reinforced Progressive Canvas Editing
arXiv CS.LG
•
ArXi:2602.01664v4 Announce Type: replace-cross In recent years, agentic workflows have been widely applied to solve complex human tasks. However, existing workflow construction still faces key challenges, including human-dependent workflow construction, the lack of graph-level execution feedback, and the inability to repair errors in-loop during long-horizon construction. To address these challenges, we propose FlowSteer, a new paradigm of Agent Designing Agentic Workflows - a single agent itself end-to-end designs the workflow that a downstream executor runs. To this paradigm, we