SkillOrchestra: Learning to Route Agents via Skill Transfer
Jiayu Wang, Yifei Ming, Zixuan Ke, Shafiq Joty, Aws Albarghouthi, Frederic Sala

TL;DR
SkillOrchestra introduces a skill-aware orchestration framework that learns fine-grained skills and models agent competence to improve routing efficiency and effectiveness in compound AI systems, outperforming RL-based methods.
Contribution
It proposes a novel skill-based approach for agent routing that reduces training costs and improves scalability and interpretability over existing RL-based orchestrators.
Findings
Outperforms state-of-the-art RL-based orchestrators by up to 22.5%.
Reduces learning costs by 700x and 300x compared to existing methods.
Demonstrates effectiveness across ten benchmark tasks.
Abstract
Compound AI systems promise capabilities beyond those of individual models, yet their success depends critically on effective orchestration. Existing routing approaches face two limitations: (1) input-level routers make coarse query-level decisions that ignore evolving task requirements; (2) RL-trained orchestrators are expensive to adapt and often suffer from routing collapse, repeatedly invoking one strong but costly option in multi-turn scenarios. We introduce SkillOrchestra, a framework for skill-aware orchestration. Instead of directly learning a routing policy end-to-end, SkillOrchestra learns fine-grained skills from execution experience and models agent-specific competence and cost under those skills. At deployment, the orchestrator infers the skill demands of the current interaction and selects agents that best satisfy them under an explicit performance-cost trade-off.…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsSoftware-Defined Networks and 5G · Advanced Graph Neural Networks · Caching and Content Delivery
