Q-RAG: Long Context Multi-step Retrieval via Value-based Embedder Training

TL;DR

Q-RAG introduces a reinforcement learning-based fine-tuning method for multi-step retrieval, enabling efficient long-context question answering with state-of-the-art results on large benchmarks.

Contribution

It presents a novel, resource-efficient approach to multi-step retrieval by training an Embedder with reinforcement learning, outperforming existing methods.

Findings

Achieves state-of-the-art results on BabiLong and RULER benchmarks.

Supports contexts up to 10 million tokens.

Offers a resource-efficient alternative to fine-tuning small LLMs.

Abstract

Retrieval-Augmented Generation (RAG) methods enhance LLM performance by efficiently filtering relevant context for LLMs, reducing hallucinations and inference cost. However, most existing RAG methods focus on single-step retrieval, which is often insufficient for answering complex questions that require multi-step search. Recently, multi-step retrieval approaches have emerged, typically involving the fine-tuning of small LLMs to perform multi-step retrieval. This type of fine-tuning is highly resource-intensive and does not enable the use of larger LLMs. In this work, we propose Q-RAG, a novel approach that fine-tunes the Embedder model for multi-step retrieval using reinforcement learning (RL). Q-RAG offers a competitive, resource-efficient alternative to existing multi-step retrieval methods for open-domain question answering and achieves state-of-the-art results on the popular long-context benchmarks BabiLong and RULER for contexts up to 10M tokens. Code is available at https://github.com/griver/Q-RAG