CapNav：面向能力条件化室内导航的视觉语言模型基准
CapNav: Benchmarking Vision Language Models on Capability-conditioned Indoor Navigation
为视觉语言模型引入「智能体物理能力」这一条件变量，并给出 45 个真实室内场景、473 个任务的能力条件化导航基准，可直接用于评测和训练 VLM 的空间尺寸推理与具身决策能力。
原文依据 CapNav provides 45 real-world indoor scenes, 473 navigation tasks, and 2365 QA pairs
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论视觉语言模型（VLM）在视觉语言导航（VLN）中已取得显著进展，为导航决策提供了新的可能性，可同时惠及机器人平台与人类用户。然而，真实世界中的导航本质上受制于智能体自身的移动能力约束。例如，扫地机器人无法通过楼梯，而四足机器人则可以。我们提出能力条件化导航（CapNav），这是一个用于评估 VLM 在给定智能体特定物理与操作能力时，能否在复杂室内空间中完成导航的基准。CapNav 定义了五种具有代表性的人类与机器人智能体，每种智能体均以物理尺寸、移动能力和环境交互能力加以描述。CapNav 提供 45 个真实室内场景、473 个导航任务和 2365 个问答对，用于检验 VLM 能否依据智能体能力在室内环境中通行。我们评估了 13 个现代 VLM，发现随着移动约束收紧，当前 VLM 的导航性能急剧下降，且即便是最优模型也难以处理需要空间尺寸推理的障碍物类型。最后，我们讨论了该结果对能力感知导航的启示，以及未来 VLM 在推进具身空间推理方面的机会。该基准已发布于 https://makeabilitylab.github.io/CapNav/
英文原文摘要 1379 字符
Vision-Language Models (VLMs) have shown remarkable progress in Vision-Language Navigation (VLN), offering new possibilities for navigation decision-making that could benefit both robotic platforms and human users. However, real-world navigation is inherently conditioned by the agent's mobility constraints. For example, a sweeping robot cannot traverse stairs, while a quadruped can. We introduce Capability-Conditioned Navigation (CapNav), a benchmark designed to evaluate how well VLMs can navigate complex indoor spaces given an agent's specific physical and operational capabilities. CapNav defines five representative human and robot agents, each described with physical dimensions, mobility capabilities, and environmental interaction abilities. CapNav provides 45 real-world indoor scenes, 473 navigation tasks, and 2365 QA pairs to test if VLMs can traverse indoor environments based on agent capabilities. We evaluate 13 modern VLMs and find that current VLM's navigation performance drops sharply as mobility constraints tighten, and that even state-of-the-art models struggle with obstacle types that require reasoning on spatial dimensions. We conclude by discussing the implications for capability-aware navigation and the opportunities for advancing embodied spatial reasoning in future VLMs. The benchmark is available at https://makeabilitylab.github.io/CapNav/