token-pruning 3 efficient-deep-learning 2 llama 2 llava 2 vision-transformer 2 computer-vision 2 token-compression 2 tome-vit 1 transformer 1 transformer-architecture 1 token-reduction 1 token-redundancy-reduction 1 token-sparsification 1 large-language-models 1 multimodal-large-language-models 1 nlp 1 token-drop 1 vicuna 1 token 1 video-transformer 1 efficient-vision-transformers 1 cvpr2024 1 model-compression 1 model-acceleration 1 vision-token-merging 1 token-fusion 1 large-vision-language-models 1 large-multimodal-models 1 efficient-inference 1