MMIT·1d Q&A: Enabling more human-like interactions and understanding in multimodal foundation models AI/ML