token-merging 9 token-pruning 5 llava 3 large-language-models 3 computer-vision 2 efficient-deep-learning 2 vision-transformer 2 token-compression 2 llama 2 token-reduction 2 multimodal-large-language-models 2 large-multimodal-models 1 efficient-inference 1 representation-learning 1 machine-learning 1 imagenet 1 image-generation 1 deep-learning 1 vicuna 1 large-vision-language-models 1 token-fusion 1 vision-token-merging 1 token-redundancy-reduction 1 token-sparsification 1 token 1 tome-vit 1 transformer 1 transformer-architecture 1 cvpr2024 1 efficient-vision-transformers 1 video-transformer 1 model-acceleration 1 model-compression 1 awq 1 benchmark 1 deepseek-v3 1 deployment 1 evaluation 1 internlm2 1 llm 1 mixtral 1 pruning 1 quantization 1 smoothquant 1 tool 1 vllm 1 wan 1 llava-next-video 1 video-large-language-models 1 visionzip 1 nlp 1 token-drop 1