Haoyuan Shi

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