一作ACL 2026Poster· Natural Language Processing
From Inheritance to Saturation: Disentangling the Evolution of Visual Redundancy for Architecture-Aware MLLM Inference Acceleration
Jiaqi Shi · Xulong Zhang · Yuechan Li · Xiaoyang Qu · Jianzong Wang
摘要
High-resolution Multimodal Large Language Models (MLLMs) face prohibitive computational costs during inference due to the explosion of visual tokens. Existing acceleration strategies, such as token pruning or layer sparsity, suffer from severe “backbone dependency”, performing well on Vicuna or M...