让大模型零成本规划：用离线检索替代实时搜索，实现快速推理。
SGA-MCTS: Decoupling Planning from Execution via Training-Free Atomic Experience Retrieval
- 将大模型规划转化为无参数检索，离线生成可复用的符号化决策单元。
- 在复杂任务上达到GPT-5级性能，无需微调且推理速度极快。
- 适合追求高效实时规划的智能系统研发者，尤其关注低成本部署。
大模型系统需具备复杂多步决策能力以应对现实任务，但现有规划方法在推理时搜索的高延迟与监督微调的泛化能力有限之间存在权衡。为此，本文提出SGA-MCTS框架，将大模型规划视为非参数检索。离线阶段，利用蒙特卡洛树搜索（MCTS）探索解空间，提炼出高保真轨迹为状态-目标-动作（SGA）原子；这些原子是去词法化的符号化单元，将具体实体抽象为符号槽位，保留可复用的因果逻辑并去除领域噪声。在线阶段，检索增强型代理采用混合符号-语义机制检索相关SGA，并将其重锚定至当前上下文作为软推理提示。实证结果表明，该范式使冻结、开放权重模型在复杂基准上达到SOTA系统（如GPT-5）的性能，无需任务特定微调。通过有效摊销搜索的高计算开销，SGA-MCTS实现了系统2式的推理深度与系统1般的推理速度，使自主规划兼具可扩展性与实时可行性。
原文摘要 · Abstract (English)
LLM-powered systems require complex multi-step decision-making abilities to solve real-world tasks, yet current planning approaches face a trade-off between the high latency of inference-time search and the limited generalization of supervised fine-tuning. To address this limitation, we introduce \textbf{SGA-MCTS}, a framework that casts LLM planning as non-parametric retrieval. Offline, we leverage Monte Carlo Tree Search (MCTS) to explore the solution space and distill high-fidelity trajectories into State-Goal-Action (SGA) atoms. These atoms are de-lexicalized primitives that abstract concrete entities into symbolic slots, preserving reusable causal logic while discarding domain-specific noise. Online, a retrieval-augmented agent employs a hybrid symbolic-semantic mechanism to fetch relevant SGAs and re-ground them into the current context as soft reasoning hints. Empirical results on complex benchmarks demonstrate that this paradigm enables frozen, open-weights models to match the performance of SOTA systems (e.g., GPT-5) without task-specific fine-tuning. By effectively amortizing the heavy computational cost of search, SGA-MCTS achieves System 2 reasoning depth at System 1 inference speeds, rendering autonomous planning both scalable and real-time feasible.
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。