CERA-MoA: Co-Evolving Routing Mechanisms with Continually Learning LLM Agents

Researchers introduced CERA-MoA, a Mixture-of-Agents framework that co-evolves query routing with continually learning LLM agents, addressing the disconnect in current MoA paradigms where routing and agent fine-tuning are treated as separate processes. The work targets the limitation that routing strategies cannot adapt to evolving agent capabilities during post-training.

Current Mixture-of-Agents MoA paradigms generally treat query routing and agent fine-tuning as separate processes, limiting their ability to respond to evolving agent capabilities. This disconnect prevents routing strategies from adapting to evolving agent capabilities during post-training and pre