Multimodal Fusion LLM

A large language model that learns joint representations across different modalities like text vision and audio in an end-to-end fashion for better cross-modal understanding and generation.