To extend the application of vision-language models (VLMs) from web images to sensor-mediated physical environments, we propose Multi-View Physical-prompt for Test-Time Adaptation (MVP), a forward-only framework that moves test-time adaptation (TTA) from tokens to photons by treating the camera exposure triangle--ISO, shutter speed, and aperture--as physical prompts. At inference, MVP acquires a library of physical views per scene, selects the top-k sensor settings using a source-affinity score, evaluates each retained view under lightweight digital augmentations, filters the lowest-entropy subset of augmented views, and aggregates predictions with Zero-temperature softmax (i.e., hard voting). This selection-then-vote design is simple, calibration-friendly, and requires no gradients or model modifications. On ImageNet-ES and ImageNet-ES-Diverse, MVP consistently outperforms digital-only TTA on single Auto-Exposure captures, by up to 25.6 percentage points (pp), and delivers up to 3.4 pp additional gains over pipelines that combine conventional sensor control with TTA. MVP remains effective under reduced parameter candidate sets that lower capture latency, demonstrating practicality. These results support the main claim that, beyond post-capture prompting, measurement-time control--selecting and combining real physical views--substantially improves robustness for VLMs.

翻译：为将视觉语言模型（VLM）的应用从网络图像扩展至传感器介导的物理环境，我们提出了多视角物理提示测试时适应（MVP），一种前向专用框架。该框架通过将相机曝光三角——ISO、快门速度和光圈——视为物理提示，将测试时适应（TTA）从标记层面推进到光子层面。在推理阶段，MVP为每个场景获取一个物理视角库，使用源亲和度分数选择前k个传感器设置，在轻量级数字增强下评估每个保留的视角，过滤出熵值最低的增强视角子集，并通过零温度softmax（即硬投票）聚合预测结果。这种“先选择后投票”的设计简洁、易于校准，且无需梯度计算或模型修改。在ImageNet-ES和ImageNet-ES-Diverse数据集上，MVP在单次自动曝光捕获场景中始终优于纯数字TTA方法，最高提升25.6个百分点（pp），并且比结合传统传感器控制与TTA的流程额外带来最高3.4 pp的性能增益。即使在减少参数候选集以降低捕获延迟的情况下，MVP仍保持有效性，证明了其实用性。这些结果支持了核心论点：除了后捕获提示，测量时控制——即选择和融合真实物理视角——能显著提升VLM的鲁棒性。