SOURCE-LINKED INTELLIGENCE

From Token Importance to Conditional Removability: Rethinking Visual Token Pruning in Multimodal Large Language Models

arXiv · AI, language, vision and robotics · article · Sep 22, 2026 · UTC

Training-free visual-token pruning often uses token importance, redundancy, or related selection criteria as proxies for safe removal. We show that these signals alone do not fully characterize removability, which is conditioned on both representation depth and the surrounding deletion set. Controlled interventions demonstrate that removing the same tokens at different depths produces substantially different downstream perturbations, while changing only the deletion context at a fixed depth alters candidate marginals and pruning-boundary decisions. These findings show that token importance alo

Read original source ↗ Open in workspace

recordType

paper

region

Global