PAS-Net is a fully multiplier-free spiking neural network that enforces human joint constraints spatially and uses causal neuromodulation temporally to achieve state-of-the-art accuracy on IMU HAR with up to 98% lower dynamic energy via early-exit.
Temporal efficient training of spiking neu- ral network via gradient re-weighting
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
years
2026 5verdicts
UNVERDICTED 5representative citing papers
SpikingMoE adds LGN-inspired SDprompt routing to a spike-driven Transformer MoE, replacing MLPs with spike-compatible experts and reporting 94.09% top-1 on CIFAR-10 and 74.54% on CIFAR-100.
LSFormer uses local structure-aware spiking self-attention and spiking response pooling to cut global attention bottlenecks, delivering 4.3% and 8.6% accuracy gains on Tiny-ImageNet and N-CALTECH101 over prior transformer-based SNNs.
Introduces circulate-firing neurons, time-step-wise learnable surrogate gradients, and balanced loss for direct SNN training, reporting competitive results on datasets and Transformers.
QDS-SNN integrates quantum neural networks with spiking neural networks via TSA-LIF neurons and QACM to report 99.72% accuracy on GTSRB in 6 steps with 55.77% lower energy than MS-ResNet baseline.
citing papers explorer
-
Towards Green Wearable Computing: A Physics-Aware Spiking Neural Network for Energy-Efficient IMU-based Human Activity Recognition
PAS-Net is a fully multiplier-free spiking neural network that enforces human joint constraints spatially and uses causal neuromodulation temporally to achieve state-of-the-art accuracy on IMU HAR with up to 98% lower dynamic energy via early-exit.
-
SpikingMoE: SDPrompt-Guided Dynamic Expert Fusion in Spiking Neural Networks
SpikingMoE adds LGN-inspired SDprompt routing to a spike-driven Transformer MoE, replacing MLPs with spike-compatible experts and reporting 94.09% top-1 on CIFAR-10 and 74.54% on CIFAR-100.
-
Breaking Global Self-Attention Bottlenecks in Transformer-based Spiking Neural Networks with Local Structure-Aware Self-Attention
LSFormer uses local structure-aware spiking self-attention and spiking response pooling to cut global attention bottlenecks, delivering 4.3% and 8.6% accuracy gains on Tiny-ImageNet and N-CALTECH101 over prior transformer-based SNNs.
-
Advancing Direct Training for Spiking Neural Networks with Circulate-Firing Neurons and Learnable Gradients
Introduces circulate-firing neurons, time-step-wise learnable surrogate gradients, and balanced loss for direct SNN training, reporting competitive results on datasets and Transformers.
-
QDS-SNN: Energy-efficient Quantum Deeply-Supervised Spiking Neural Network Algorithm for Traffic Sign Recognition
QDS-SNN integrates quantum neural networks with spiking neural networks via TSA-LIF neurons and QACM to report 99.72% accuracy on GTSRB in 6 steps with 55.77% lower energy than MS-ResNet baseline.