Research map / Architectures and efficiency / Spiking networks
Parallel Spiking Neurons with High Efficiency and Ability to Learn Long-term Dependencies
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use PSN with fully connected temporal weights instead of conventional spiking neurons | SNNs · Sequential CIFAR10 | test accuracy (%): PSN 88.45; LIF 81.50 | Helped |
| Use PSN with fully connected temporal weights instead of conventional spiking neurons | SNNs · Sequential CIFAR100 | test accuracy (%): PSN 62.21; LIF 55.45 | Helped |
| Add high-order temporal connections using masked PSN | SNNs · Sequential CIFAR10 | test accuracy (%): masked PSN 85.81; LIF 81.50 | Helped |
| Share masked PSN weights across time-steps with sliding PSN | SNNs · Sequential CIFAR100 | test accuracy (%): sliding PSN 62.11; masked PSN 60.69 | Helped |
| Remove reset from vanilla LIF dynamics (ablation) | LIF · Sequential CIFAR10 | test accuracy (%): 81.50 -> 79.50 | Hurt |
| Remove reset from vanilla LIF dynamics (ablation) | LIF · Sequential CIFAR100 | test accuracy (%): 55.45 -> 53.33 | Hurt |
| Increase masked PSN order from 1 to 4 (ablation) | masked PSN · Sequential CIFAR100 | accuracy | Helped |
| Increase PSN order beyond 4 (ablation) | masked PSN and sliding PSN · Sequential CIFAR100 | accuracy | Mixed |
| Use a non-largest order (20 for masked PSN; 31 for sliding PSN) (ablation) | masked PSN and sliding PSN · Sequential CIFAR100 | accuracy | Helped |
| Replace SEW ResNet neurons with PSN | SEW ResNet-34 · ImageNet | accuracy (%): 67.04 -> 70.54 | Helped |
| Use sliding PSN (k=2) instead of reported prior methods | VGG · CIFAR10-DVS | accuracy (%): 85.30%; 2+% higher than the previous SOTA method | Helped |
Closest papers by meaning
- Multiplication-Free Parallelizable Spiking Neurons with Efficient Spatio-Temporal Dynamics (NeurIPS 2025 · similarity 0.91)
- Efficient Parallel Training Methods for Spiking Neural Networks with Constant Time Complexity (ICML 2025 · similarity 0.90)
- SpikingSSMs: Learning Long Sequences with Sparse and Parallel Spiking State Space Models (AAAI 2025 · similarity 0.89)
- The Dormant Spiking Neuron: A State-Driven Mechanism for Efficient Spiking Neural Networks (NeurIPS 2026 · similarity 0.88)
- Faster and Stronger: When ANN-SNN Conversion Meets Parallel Spiking Calculation (ICML 2025 · similarity 0.88)
- P-SPIKESSM: HARNESSING PROBABILISTIC SPIKING STATE SPACE MODELS FOR LONG-RANGE DEPENDENCY TASKS (ICLR 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.