A variance-based, retraining-free pruning framework for vision-language models that allocates per-layer sparsity and outperforms Wanda and SparseGPT at high sparsity.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Prune Once: Retraining-Free Task-Agnostic Pruning for Vision-Language Models
A variance-based, retraining-free pruning framework for vision-language models that allocates per-layer sparsity and outperforms Wanda and SparseGPT at high sparsity.