MCTS-RAG：利用蒙特卡洛樹搜索增強檢索增強生成

MCTS-RAG: Enhancing Retrieval-Augmented Generation with Monte Carlo Tree Search

March 26, 2025

作者: Yunhai Hu, Yilun Zhao, Chen Zhao, Arman Cohan

cs.AI

摘要

我們提出了MCTS-RAG，這是一種新穎的方法，通過利用檢索增強生成（RAG）來提供相關上下文，並結合蒙特卡羅樹搜索（MCTS）來精煉推理路徑，從而增強小型語言模型在知識密集型任務中的推理能力。MCTS-RAG通過迭代的決策過程動態整合檢索與推理。與標準的RAG方法（通常獨立於推理進行信息檢索，從而導致知識整合不理想）或傳統的MCTS推理（僅依賴內部模型知識而無外部事實）不同，MCTS-RAG將結構化推理與自適應檢索相結合。這種整合方法增強了決策能力，減少了幻覺，並確保了更高的事實準確性和回答一致性。在多個推理和知識密集型數據集（如ComplexWebQA、GPQA和FoolMeTwice）上的實驗結果表明，我們的方法使小型語言模型能夠通過有效擴展推理時的計算資源，達到與GPT-4o等前沿大型語言模型相當的性能，為小型模型的推理樹立了新標準。

English

We introduce MCTS-RAG, a novel approach that enhances the reasoning capabilities of small language models on knowledge-intensive tasks by leveraging retrieval-augmented generation (RAG) to provide relevant context and Monte Carlo Tree Search (MCTS) to refine reasoning paths. MCTS-RAG dynamically integrates retrieval and reasoning through an iterative decision-making process. Unlike standard RAG methods, which typically retrieve information independently from reasoning and thus integrate knowledge suboptimally, or conventional MCTS reasoning, which depends solely on internal model knowledge without external facts, MCTS-RAG combines structured reasoning with adaptive retrieval. This integrated approach enhances decision-making, reduces hallucinations, and ensures improved factual accuracy and response consistency. The experimental results on multiple reasoning and knowledge-intensive datasets datasets (i.e., ComplexWebQA, GPQA, and FoolMeTwice) show that our method enables small-scale LMs to achieve performance comparable to frontier LLMs like GPT-4o by effectively scaling inference-time compute, setting a new standard for reasoning in small-scale models.