CLSE prunes tokens in MLLMs by quantifying cross-layer spectral redistribution in the frequency domain to preserve semantically active tokens and reduce compute.
arXiv preprint arXiv:2410.07278 (2024)
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
PriorTR estimates model-induced prior attention via a null token in one forward pass and contrasts it with task-conditioned attention to improve visual token pruning accuracy-efficiency trade-offs in MLLMs.
citing papers explorer
-
Spectral Evolution-Guided Token Pruning in Multimodal Large Language Models
CLSE prunes tokens in MLLMs by quantifying cross-layer spectral redistribution in the frequency domain to preserve semantically active tokens and reduce compute.
-
Accelerating Multimodal Large Language Models with Prior-Corrected Token Reduction
PriorTR estimates model-induced prior attention via a null token in one forward pass and contrasts it with task-conditioned attention to improve visual token pruning accuracy-efficiency trade-offs in MLLMs.