Rethinking Retrieval-Augmented Generation as a Cooperative Decision-Making Problem

TL;DR

This paper introduces CoRAG, a cooperative framework for retrieval-augmented generation that jointly optimizes rerankers and generators as peer decision-makers, improving stability and generalization.

Contribution

It proposes a novel cooperative approach to RAG, treating reranker and generator as peers to enhance collaboration and performance.

Findings

CoRAG improves generation stability over traditional RAG models.

The model generalizes well with limited training data (~10K samples).

Experimental results show enhanced cooperation leads to better responses.

Abstract

Retrieval-Augmented Generation (RAG) has demonstrated strong effectiveness in knowledge-intensive tasks by grounding language generation in external evidence. Despite its success, many existing RAG systems are built based on a ranking-centric, asymmetric dependency paradigm, where the generation quality of the generator is highly dependent on reranking results of the reranker. To overcome this limitation, we propose Cooperative Retrieval-Augmented Generation (CoRAG), a framework that treats the reranker and the generator as peer decision-makers rather than being connected through an asymmetric dependency pipeline. By jointly optimizing their behaviors toward a shared task objective, the reranker and generator are encouraged to cooperate, ensuring that document reranking and generation work in concert to improve the final response. Experimental results demonstrate good generalization and improved generation stability of CoRAG, even when the model is trained on only around 10K PopQA samples. Our model released in https://github.com/CoderrrSong/CoRAG.