CLSE prunes tokens in MLLMs by quantifying cross-layer spectral redistribution in the frequency domain to preserve semantically active tokens and reduce compute.
Harsh Jhamtani and Taylor Berg-Kirkpatrick
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
years
2026 4verdicts
UNVERDICTED 4representative 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.
CIVIC is a path-consistent compact visual inference framework that reduces KV-cache memory to approximately one-third and end-to-end latency in VLMs while preserving accuracy via text-aligned KL distillation and adaptive spatial retention.
HybridKV reduces KV cache memory by up to 7.9x and speeds decoding by 1.52x in MLLMs with almost no performance loss by classifying heads into static and dynamic types and compressing them differently.
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.
-
CIVIC: End-to-End Sequence Compactness for Efficient Vision-Language Models
CIVIC is a path-consistent compact visual inference framework that reduces KV-cache memory to approximately one-third and end-to-end latency in VLMs while preserving accuracy via text-aligned KL distillation and adaptive spatial retention.
-
HybridKV: Hybrid KV Cache Compression for Efficient Multimodal Large Language Model Inference
HybridKV reduces KV cache memory by up to 7.9x and speeds decoding by 1.52x in MLLMs with almost no performance loss by classifying heads into static and dynamic types and compressing them differently.