DEALS: Decentralized Expertise-Aware Load Serving for Multi-Agent LLM Systems
Experiments show that DEALS not only improves performance along multiple dimensions (e.g., answer accuracy and task throughput) in both homogeneous and heterogeneous agent pools, but also balances agent expertise and workload in a self-organized manner, enabling effective decentralized coordination.
Jing-Hang Huang, Wen-Bin Wang, Yan-Chuan Yin et al.
· 0 citations