Yi-Cheng Qiu

A real-time multimodal digital human interaction system via cross-attention fusion of LLaMA and CLIP models

The main contribution is a token-level cross-attention module that aligns CLIP visual features with LLaMA language states and an interest-decay LSTM that updates user profiles after each interaction that improves multimodal matching, intention recognition, and personalized recommendation while retaining real-time infer...