SmartRAG: Jointly Learn RAG-Related Tasks From the Environment Feedback
This work addresses inefficiencies in RAG systems for AI applications by enabling joint optimization, though it is incremental as it builds on existing RAG frameworks.
The authors tackled the problem of separately trained modules in RAG systems by proposing SmartRAG, a pipeline with a policy network and retriever that are jointly optimized using reinforcement learning to improve performance and reduce retrieval costs, achieving better results than separately optimized versions.
RAG systems consist of multiple modules to work together. However, these modules are usually separately trained. We argue that a system like RAG that incorporates multiple modules should be jointly optimized to achieve optimal performance. To demonstrate this, we design a specific pipeline called \textbf{SmartRAG} that includes a policy network and a retriever. The policy network can serve as 1) a decision maker that decides when to retrieve, 2) a query rewriter to generate a query most suited to the retriever, and 3) an answer generator that produces the final response with/without the observations. We then propose to jointly optimize the whole system using a reinforcement learning algorithm, with the reward designed to encourage the system to achieve the best performance with minimal retrieval cost. When jointly optimized, all the modules can be aware of how other modules are working and thus find the best way to work together as a complete system. Empirical results demonstrate that the jointly optimized SmartRAG can achieve better performance than separately optimized counterparts.