SOURCE-LINKED INTELLIGENCE

Layer-Aware Position Embeddings for Visual Token Pruning in Multimodal Large Language Models

arXiv · AI, language, vision and robotics · article · Sep 20, 2026 · UTC

Multimodal large language models (MLLMs) incur substantial computational overhead due to the reliance on hundreds of visual tokens to represent images. While token pruning has emerged as a promising approach to reduce the inference cost of MLLMs, existing methods typically reassign position embeddings to the retained tokens using either sparse or continuous position embeddings, each introducing distinct limitations. Sparse position embeddings tend to decrease the attention value allocated to visual tokens, thereby degrading the perception capability of MLLMs, whereas continuous position embedd

Read original source ↗ Open in workspace

recordType

paper

region

Global