PhysVLM-AVR: Active Visual Reasoning for Multimodal Large Language Models in Physical Environments | Read Paper on Bytez