Yi Zhang

This paper is to jointly optimize mutiple modules in complex RAG pipeline using multi-agent reinforcement learning.