The massive redundancy of visual tokens in Vision-Language Models (VLMs) creates a critical computational bottleneck during large language model (LLM) prefill. Recent training-free pruning methods provide plug-and-play acceleration, yet their pruning decisions often rely on limited visual states or delayed in-LLM signa...
Qing-Xin Liu, Jin-Zhi Liao, Xiang Zhao· Proceedings of the 35th ACM...· 0 citations
The massive redundancy of visual tokens in Vision-Language Models (VLMs) creates a critical computational bottleneck during large language model (LLM) prefill. Recent training-free pruning methods provide plug-and-play acceleration, yet their pruning decisions often rely on limited visual states or delayed in-LLM signa...
Qing-Xin Liu, Jin-Zhi Liao, Xiang Zhao· Proceedings of the 35th ACM...· 0 citations
A neuron-level controllable ICE framework for MKE, namely NeuCon-ICE, which consists of a multimodal contextual neuron identification module that aims to determine where to edit by identifying the tightly coupled multimodal contextual neurons, and a context-aware neuron editing module that aims to determine how to edit...