AI RESEARCH

Is VLA Reasoning Faithful? Probing Safety of Chain-of-Causation

arXiv CS.CV

ArXi:2605.17268v1 Announce Type: cross We present the first systematic study of faithfulness in Vision-Language-Action (VLA) driving models, analyzing 300 Alpamayo-R1-10B inferences across 100 diverse PhysicalAI-AV scenarios.