发表机构
The University of Texas at Austin; Collins Aerospace（德克萨斯大学奥斯汀分校; 柯林斯航空航天公司）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对视觉-语言导航智能体的条件分支诊断需求，提出基于场景图的基准CondVLN及轻量级神经符号分支选择模型，可暴露智能体的逻辑决策失败并提升性能2倍。
AI 中文摘要
视觉-语言导航智能体通常会依据类路线指令朝着固定目标的能力进行评估，但现实中的导航指令往往依赖于对环境观测状态的判断：若某一条件成立则遵循一条路径，否则选择另一条路径。这类指令要求智能体评估场景证据、选择正确的逻辑分支并执行对应的导航行为。现有评估对条件分支执行的控制有限，难以确定智能体的失败是源于感知、 grounding（语义关联）、导航还是逻辑决策。我们提出CondVLN，这是一个基于场景图的、用于诊断视觉-语言导航中条件分支的基准。CondVLN通过程序生成指令，其分支条件基于可验证的3D场景图谓词，且在分支深度、依赖链长度、空间组成、证据可观测性及指令范围上有可控变化。CondVLN在AI2-THOR、Matterport3D、Gibson和ReplicaCAD中包含超过11500条生成的条件指令，并使用标准VLN指标及分支特定诊断指标（分支选择准确率和条件成功率）评估智能体。对四个最先进的VLN智能体（VLN-Zero、NaVid、NaVILA和Open-Nav）的评估显示，条件分支会暴露出标准成功率或路径长度无法捕捉的失败：智能体可能在合理导航的同时选择与观测场景条件不一致的分支。我们还提出了一个轻量级的神经符号分支选择模型，该模型将条件 grounding 与导航执行分离，使性能提升2倍。CondVLN提供了一个可复用的测试平台，用于评估具身智能体是否不仅能遵循指令，还能在正确条件下遵循正确的指令。
英文摘要
Vision-language navigation agents are often evaluated on their ability to follow route-like instructions toward a fixed goal. Yet, real navigation instructions often depend on observed states of the environment: if a condition holds, then follow one path, otherwise take another. Such instructions require an agent to evaluate scene evidence, select the correct logical branch, and execute the corresponding navigation behavior. Existing evaluations provide limited control over conditional branch execution, making it difficult to determine whether agents fail because of perception, grounding, navigation, or logical decision-making. We introduce CondVLN, a scene-graph-grounded benchmark for diagnosing conditional branching in vision-language navigation. CondVLN programmatically generates instructions whose branch conditions are grounded in verifiable 3D scene-graph predicates, with controlled variation in branch depth, dependency chain length, spatial composition, evidence observability, and instruction horizon. CondVLN contains over 11,500 generated conditional instructions across AI2-THOR, Matterport3D, Gibson, and ReplicaCAD, and evaluates agents using standard VLN metrics and branch-specific diagnostics: Branch Selection Accuracy and Conditional Success Rate. Evaluating four state-of-the-art VLN agents (VLN-Zero, NaVid, NaVILA, and Open-Nav) shows that conditional branching exposes failures that are not captured by standard success rate or path length alone: agents can navigate plausibly while committing to a branch inconsistent with the observed scene condition. We also present a lightweight neurosymbolic branch-selection model that separates condition grounding from navigation execution, improving performance by 2x. CondVLN provides a reusable testbed for measuring whether embodied agents can not only follow instructions, but follow the right instruction under the right condition.
Comments11 pages, 1 figure, 3 tables. Project page: https://condvln.github.io/
