通过复用过往推理步骤，让大模型更快更省地思考。
Retrieval-of-Thought: Efficient Reasoning via Reusing Thoughts
- 将已有推理过程组织成可检索的思维图谱，支持快速调用与重组。
- 在多个模型上实现最多40%输出token减少、82%延迟降低和59%成本下降。
- 适合追求高效推理且希望保持高准确率的研究者与工程师。
大型推理模型通过生成长推理轨迹提升准确率，但带来延迟与成本上升。本文提出检索式思维（Retrieval-of-Thought, RoT），通过复用以往推理中的可组合‘思维’步骤来引导新问题求解。RoT将思维步骤构建成包含顺序与语义边的思维图谱，支持快速检索与灵活重组。推理时，系统检索相关节点，并采用奖励引导的遍历策略构建针对具体问题的动态模板以指导生成。该方法减少了冗余探索，从而降低输出token数量，同时保持准确率。在多个模型与推理基准上的评估显示，尽管提示词增长微小，但效率显著提升：输出token减少最多达40%，推理延迟降低82%，成本下降59%，准确率不变。RoT建立了一种通过检索实现动态模板构建的高效大模型推理可扩展范式。
原文摘要 · Abstract (English)
Large reasoning models improve accuracy by producing long reasoning traces, but this inflates latency and cost, motivating inference-time efficiency. We propose Retrieval-of-Thought (RoT), which reuses prior reasoning as composable ``thought" steps to guide new problems. RoT organizes steps into a thought graph with sequential and semantic edges to enable fast retrieval and flexible recombination. At inference, RoT retrieves query-relevant nodes and applies reward-guided traversal to assemble a problem-specific template that guides generation. This dynamic template reuse reduces redundant exploration and, therefore, reduces output tokens while preserving accuracy. We evaluate RoT on reasoning benchmarks with multiple models, measuring accuracy, token usage, latency, and memory overhead. Findings show small prompt growth but substantial efficiency gains, with RoT reducing output tokens by up to 40%, inference latency by 82%, and cost by 59% while maintaining accuracy. RoT establishes a scalable paradigm for efficient LRM reasoning via dynamic template construction through retrieval.
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。