Multi-SpatialMLLM: Multi-Frame Spatial Understanding with Multi-Modal Large Language Models | Read Paper on Bytez