Centering before Pruning: Lightweight Geometry Correction for Diversity-Based Visual Token Pruning in LVLMs

2026-08-31

AI Summary

本文针对LVLMs中冗余视觉token序列导致的高推理成本问题，提出了一种基于中心化几何修正的方法Cen-Prune来优化多样性修剪过程。

Abstract

Large vision-language models (LVLMs) incur substantial inference costs due to their long and highly redundant visual-token sequences. Diversity-based pruning mitigates this cost by selecting token subsets based on pairwise cosine similarity. We find, however, that similarities between raw visual tokens are strongly concentrated in the positive range, limiting their ability to distinguish non-redundant tokens. A natural way to improve this resolution is to center token features before computing cosine similarity. Centering indeed reveals a substantially richer pairwise structure, yet unexpectedly degrades pruning performance when used alone. We show that this apparent contradiction arises because the raw geometry does more than represent pairwise diversity: it also implicitly favors globally distinctive tokens, which tend to contain semantically informative content. Centering better resolves subset diversity but loses this useful token-wise preference, revealing that diversity and distinctiveness are entangled in the raw geometry. Based on this analysis, we propose the \textbf{Cen}tered Geometry \textbf{Prune}r (Cen-Prune), which measures subset diversity using centered cosine similarity while retaining raw-space distinctiveness as a complementary token-wise preference. This lightweight, plug-and-play correction leaves the underlying selection mechanism unchanged and incurs negligible computational overhead. Extensive experiments across multiple image- and video-understanding benchmarks and LVLM architectures demonstrate that Cen-Prune provides robust improvements in overall performance across existing diversity-based pruners.

Problem

• visual-token pruning
• diversity-based pruning
• cosine similarity
• centering
• distinctiveness

Innovation

• centering
• diversity-based pruning
• visual tokens
• cosine similarity
• LVLMs