Distilling Implicit Multimodal Knowledge into Large Language Models for Zero-Resource Dialogue Generation

VLM
Re-assign community