用蒙特卡洛树搜索优化小模型的推理，提升事实准确性。
MCTS-RAG: Enhancing Retrieval-Augmented Generation with Monte Carlo Tree Search
- 结合检索与树搜索，动态调整推理路径
- 在复杂问答数据集上接近GPT-4o表现
- 适合资源有限但需高准确推理的场景
我们提出MCTS-RAG，通过检索增强生成（RAG）提供相关上下文，并利用蒙特卡洛树搜索（MCTS）优化推理路径，增强小语言模型在知识密集型任务中的推理能力。该方法通过迭代决策过程动态融合检索与推理，克服了传统RAG独立于推理导致知识整合不佳，以及常规MCTS仅依赖内部知识而缺乏外部事实的缺陷。MCTS-RAG将结构化推理与自适应检索结合，提升决策质量，减少幻觉，保证事实准确性和响应一致性。在ComplexWebQA、GPQA和FoolMeTwice等多个推理与知识密集型数据集上的实验表明，该方法使小规模语言模型在有效扩展推理时计算资源后，性能可媲美GPT-4o等前沿大模型，为小模型推理树立新标准。
原文摘要 · Abstract (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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。