AI Summary: This paper tackles the escalating computational costs of running multi-agent systems by introducing a 'Difficulty-Aware' routing mechanism. Instead of pushing every user prompt through a complex, multi-step agent debate or verification loop, DAAO uses a lightweight classifier to predict how hard the query is.
Existing multi-agent frameworks often rely on static or task-level workflows, which either over-process simple queries or underperform on complex ones, while neglecting the efficiency-performance trade-offs across heterogeneous LLMs. To address these limitations, we propose Difficulty-Aware Agentic Orchestration (DAAO), which dynamically generates query-specific workflows guided by predicted query difficulty. A self-adjusting policy updates difficulty estimates based on workflow success, enabling simpler workflows for easy queries and more complex strategies for harder ones. Experiments on six benchmarks demonstrate that DAAO surpasses prior multi-agent systems in both accuracy and inference efficiency.
Share your opinion to help other learners triage faster.
Write a reviewInvite someone by email to share an invited review for Difficulty-Aware Agentic Orchestration for Query-Specific Multi-Agent Workflows.