TCP-SSM conditions stable poles on visual tokens to explicitly control memory decay and oscillation in SSMs, cutting computation up to 44% while matching or exceeding accuracy on classification, segmentation, and detection.
Dynamic vision mamba
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
Vision SmolMamba adds spike-guided spatio-temporal token pruning to a bidirectional spiking state-space backbone, cutting estimated energy by at least 1.5x versus prior spiking Transformers and Spiking Mamba variants on ImageNet-1K and event-based datasets while keeping competitive accuracy.
citing papers explorer
-
TCP-SSM: Efficient Vision State Space Models with Token-Conditioned Poles
TCP-SSM conditions stable poles on visual tokens to explicitly control memory decay and oscillation in SSMs, cutting computation up to 44% while matching or exceeding accuracy on classification, segmentation, and detection.
-
Vision SmolMamba: Spike-Guided Token Pruning for Energy-Efficient Spiking State-Space Vision Models
Vision SmolMamba adds spike-guided spatio-temporal token pruning to a bidirectional spiking state-space backbone, cutting estimated energy by at least 1.5x versus prior spiking Transformers and Spiking Mamba variants on ImageNet-1K and event-based datasets while keeping competitive accuracy.