Haoyuan Shi

model compression (1) neural network pruning (1) benchmark evaluation (1) iterative optimization (1) visual question answering (1) self-supervised learning (1) knowledge distillation (1) multimodal learning (1) electron microscopy (1) sparse model (1) mamba architecture (1) large multimodal model (1) token prediction (1) multimodal comprehension (1) multimodal model (1) cultural diversity (1) visual knowledge alignment (1) knowledge alignment (1) vision language (1) large language model (1)