Embodied4C: 评估具身视觉-语言导航中至关重要的因素
Embodied4C: Measuring What Matters for Embodied Vision-Language Navigation
AI总结：
Embodied4C通过三种异构具身评估VLMs的具身能力，发现跨模态对齐和指令微调比规模更重要，而空间和时间推理是可靠具身能力的主要瓶颈。
摘要：
视觉-语言导航要求智能体在具身约束下进行推理和行动。尽管视觉-语言模型（VLMs）表现出强大的泛化能力，但当前的基准测试对具身——即物理平台的选择、传感器配置和模态对齐——如何影响感知、推理和控制的理解有限。我们介绍了Embodied4C，一个闭环基准测试，旨在作为具身推理的图灵测试。该基准测试通过大约1100个单次推理问题和58个目标导向的导航任务，评估VLMs在三种异构具身——自动驾驶汽车、无人机和机械臂——中的核心具身能力。这些任务共同评估四个基础维度：语义、空间、时间和物理推理。每种具身都提供动态传感器配置和环境变化，以测试超出平台特定适应的泛化能力。为防止具身过拟合，Embodied4C整合了跨领域查询，针对抽象和跨情境推理。对十种最先进的VLMs和四种具身控制基线的全面评估显示，跨模态对齐和指令微调比规模更重要，而空间和时间推理仍然是可靠具身能力的主要瓶颈。
英文摘要：
Vision-language navigation requires agents to reason and act under constraints of embodiment. While vision-language models (VLMs) demonstrate strong generalization, current benchmarks provide limited understanding of how embodiment -- i.e., the choice of physical platform, sensor configuration, and modality alignment -- influences perception, reasoning, and control. We introduce Embodied4C, a closed-loop benchmark designed as a Turing test for embodied reasoning. The benchmark evaluates the core embodied capabilities of VLMs across three heterogeneous embodiments -- autonomous vehicles, aerial drones, and robotic manipulators -- through approximately 1.1K one-shot reasoning questions and 58 goal-directed navigation tasks. These tasks jointly assess four foundational dimensions: semantic, spatial, temporal, and physical reasoning. Each embodiment presents dynamic sensor configurations and environment variations to probe generalization beyond platform-specific adaptation. To prevent embodiment overfitting, Embodied4C integrates domain-far queries targeting abstract and cross-context reasoning. Comprehensive evaluation across ten state-of-the-art VLMs and four embodied control baselines shows that cross-modal alignment and instruction tuning matter more than scale, while spatial and temporal reasoning remains the primary bottleneck for reliable embodied competence.
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。
