FREESON: Retriever-Free Retrieval-Augmented Reasoning via Corpus-Traversing MCTS

TL;DR

FREESON introduces a novel retriever-free reasoning framework using corpus-traversing MCTS, enabling large reasoning models to independently locate relevant knowledge within the corpus, improving efficiency and accuracy in multi-step question answering.

Contribution

This work presents a new retrieval-augmented reasoning approach that eliminates the need for separate retrievers by integrating corpus traversal into the reasoning process with MCTS, enhancing performance and reducing costs.

Findings

Achieves 14.4% average improvement in EM and F1 over models with separate retrievers.

Performs comparably or better than strong baselines on five open-domain QA benchmarks.

Effectively handles both single-hop and multi-hop questions with improved accuracy.

Abstract

Large Reasoning Models (LRMs) have demonstrated remarkable capabilities in multi-step reasoning and calling search engines at appropriate steps. However, existing retrieval-augmented reasoning approaches rely on separate retrieval models, limiting the LRM's role in retrieval to deciding when to retrieve and how to query. This separation not only increases hardware and operational costs but also leads to errors in the retrieval process due to the representation bottleneck, a phenomenon where the retriever's embedding space is not expressive enough to meet the generator's requirements. To address this, we shift our perspective from sequence-to-sequence matching to locating the answer-containing paths within the corpus, and propose a novel framework called FREESON (Retriever-FREE Retrieval-Augmented ReaSONing). This framework enables LRMs to retrieve relevant knowledge on their own by acting as both a generator and retriever. To achieve this, we introduce a variant of the MCTS algorithm specialized for the retrieval task, which we call CT-MCTS (Corpus-Traversing Monte Carlo Tree Search). In this algorithm, LRMs traverse through the corpus toward answer-containing regions. Our results on five open-domain QA benchmarks, including single-hop and multi-hop questions, show that FREESON achieves an average improvement of 14.4% in EM and F1 over four multi-step reasoning models with a separate retriever, and it also performs comparably to the strongest baseline, surpassing it by 3% on PopQA and 2WikiMultihopQA.