Thomas Stauner

Finn Rasmus Schäfer, Yuan Gao, Dingrui Wang et al.

While Vision-Language Models (VLMs) have advanced highlevel reasoning in autonomous driving, their ability to ground this reasoning in the underlying physics of ego-motion remains poorly understood. We introduce EgoDyn-Bench, a diagnostic benchmark for evaluating the semantic ego-motion understanding of vision-centric foundation models. By mapping continuous vehicle kinematics to discrete motion concepts via a deterministic oracle, we decouple a model's internal physical logic from its visual perception. Our large-scale empirical audit spanning 20 + models, including closed-source MLLMs, open-source VLMs across multiple scales, and specialized VLAs, identifies a significant Perception Bottleneck: while models exhibit logical physical concepts, they consistently fail to accurately align them with visual observations, frequently underperforming classical non-learned geometric baselines. This failure persists across model scales and domain-specific training, indicating a structural deficit in how current architectures couple visual perception with physical reasoning. We demonstrate that providing explicit trajectory encodings substantially restores physical consistency across all evaluated models, revealing a functional disentanglement between vision and language: egomotion logic is derived almost exclusively from the language modality, while visual observations contribute negligible additional signal. This structural finding provides a standardized diagnostic framework and a practical pathway toward physically aligned embodied AI. Keywords: Ego-motion - Physical Reasoning - Foundation Models

Alexander Pretschner, Wolfgang Prenninger, Stefan Wagner et al.

Model-based testing relies on behavior models for the generation of model traces: input and expected output---test cases---for an implementation. We use the case study of an automotive network controller to assess different test suites in terms of error detection, model coverage, and implementation coverage. Some of these suites were generated automatically with and without models, purely at random, and with dedicated functional test selection criteria. Other suites were derived manually, with and without the model at hand. Both automatically and manually derived model-based test suites detected significantly more requirements errors than hand-crafted test suites that were directly derived from the requirements. The number of detected programming errors did not depend on the use of models. Automatically generated model-based test suites detected as many errors as hand-crafted model-based suites with the same number of tests. A sixfold increase in the number of model-based tests led to an 11% increase in detected errors.