基于部分可观测性的文本基于探索与导航中的大型语言模型
LLMs for Text-Based Exploration and Navigation Under Partial Observability
AI总结：
本文研究了大型语言模型在部分可观测环境下作为纯文本控制器的可行性，评估了九种不同LLM在探索和导航任务中的表现，发现推理调优模型在导航任务中表现更可靠，但效率仍低于理想路径。
摘要：
探索和目标导向的导航在未知布局中是检查、物流和搜索与救援的关键。我们询问大型语言模型（LLMs）是否可以在部分可观测性下作为纯文本控制器工作，而无需代码执行、工具或程序合成。我们引入了一个可重复的基准测试，包含固定ASCII网格世界中的oracle本地化：每一步只揭示agent周围局部5×5的窗口，模型必须选择UP/RIGHT/DOWN/LEFT之一。九种当代LLM，从开放/专有、密集/Mixture of Experts和指令-推理调优的模型，在三个难度递增的布局上进行了两项任务的评估：探索（最大化揭示的单元格）和导航（到达目标的最短路径）。实验结果在成功率、效率（如归一化覆盖）和路径长度等定量指标以及定性分析上进行了评估。推理调优模型在所有布局中可靠地完成了导航，但效率仍低于理想路径。提示中的少样本演示主要帮助这些推理调优模型，减少无效移动并缩短路径，而经典密集指令模型仍然不一致。我们观察到具有UP/RIGHT特征动作先验，这可以导致部分可观测性下的循环。总体而言，训练制度和测试时的推理预测控制能力比原始参数数量更好。这些发现表明，轻量级混合与经典在线规划器的结合是部署部分地图系统的一种实用路线。
英文摘要：
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.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
