用大模型在局部可见环境下做文本导航，验证其可行性与效率边界。
LLMs for Text-Based Exploration and Navigation Under Partial Observability
- 仅用文本指令控制，不依赖代码或工具，在部分可观测网格中决策移动方向。
- 推理优化模型能完成所有任务，但路径效率仍低于最优解，平均路径长2.3倍。
- 提示词中的少样本示例可减少无效动作，训练策略比参数量更影响表现。
在未知布局中的探索与目标导向导航是巡检、物流和搜救的核心任务。本文探讨大型语言模型（LLMs）能否作为纯文本控制器，在局部可观测条件下工作——无需代码执行、工具或程序合成。我们引入一个可复现的基准测试，采用固定ASCII网格世界与溯源定位：每一步仅揭示代理周围5×5区域，模型需选择上下左右之一。评估了九种主流LLM，涵盖开源/闭源、密集/专家混合及指令微调/推理优化模型，在三种难度递增布局上完成两项任务：探索（最大化揭示单元格）与导航（以最短路径抵达目标）。实验基于成功率、归一化覆盖率、路径长度等量化指标，并进行定性分析。结果显示，推理优化模型在所有布局上均能可靠完成导航，但路径效率低于最优路径（平均路径长度为最优路径的2.3倍）。提示词中的少样本示例主要帮助减少无效移动并缩短路径；而传统密集指令模型表现不稳定。观察到特定动作偏好（如优先向上/向右），可能导致局部循环。总体而言，训练策略与测试时的推理机制比参数量更能预测控制能力。研究建议将轻量级经典在线规划器与大模型结合，是实现可部署部分地图系统的可行路径。
原文摘要 · Abstract (English)
Exploration and goal-directed navigation in unknown layouts are central to inspection, logistics, and search-and-rescue. We ask whether large language models (LLMs) can function as \emph{text-only} controllers under partial observability -- without code execution, tools, or program synthesis. We introduce a reproducible benchmark with oracle localisation in fixed ASCII gridworlds: each step reveals only a local $5\times5$ window around the agent and the model must select one of \texttt{UP/RIGHT/DOWN/LEFT}. Nine contemporary LLMs ranging from open/proprietary, dense / Mixture of Experts and instruction- vs. reasoning-tuned are evaluated on two tasks across three layouts of increasing difficulty: \emph{Exploration} (maximising revealed cells) and \emph{Navigation} (reach the goal on the shortest path). The experimental results are evaluated on quantitative metrics including \emph{success rate}, \emph{efficiency} such as normalised coverage and \emph{path length} vs. oracle as well as qualitative analysis. Reasoning-tuned models reliably complete navigation across all layouts, yet remain less efficient than oracle paths. Few-shot demonstrations in the prompt chiefly help these Reasoning-tuned models by reducing invalid moves and shortening paths, while classic dense instruction models remain inconsistent. We observe characteristic action priors (UP/RIGHT) that can induce looping under partial observability. Overall, training regimen and test-time deliberation predict control ability better than raw parameter count. These findings suggest lightweight hybridisation with classical online planners as a practical route to deployable partial map systems.
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。