SOURCE-LINKED INTELLIGENCE

Cut-ViT: Task-Specific Model Pruning via Gram Anchoring Subspace Consistency

arXiv · AI, language, vision and robotics · article · Aug 28, 2026 · UTC

Pruning visual foundation models has attracted considerable attention. However, existing methods focus on rigid point-to-point token alignment on a single dataset for pruning, suffering from two limitations: i) robustness degradation, and ii) task-specificity deficiency. To address these limitations, we propose a task-specific pruning pipeline, named Cut-ViT. Specifically, we first construct gram anchoring matrices from both spatial and semantic perspectives, and perform the subspace decomposition to extract the corresponding subspace bases. Basis-agnostic and residual constraints are then ado

Read original source ↗ Open in workspace

recordType

paper

region

Global