TAMP: Token-Adaptive Layerwise Pruning in Multimodal Large Language Models

VLM
Re-assign community