What do MLLMs hear? Examining reasoning with text and sound components in Multimodal Large Language Models

Open in new window