AI Paper Atlas

Research map / Architectures and efficiency / Spiking networks

SpikePack: Enhanced Information Flow in Spiking Neural Networks with High Hardware Compatibility

Guobin Shen, Jindong Li, Tenglong Li, Dongcheng Zhao, Yi Zeng

ICCV 2025 · Architectures and efficiency › Spiking networks · approach: Training deep spiking networks · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (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 changedWhereMeasuredEffect
replace LIF neurons with SpikePack in direct trainingResNet-34 (T=4) · ImageNet-1ktop-1 accuracy (%): SpikePack 73.4 vs T-RevSNN 73.2 / TET 68.0 / SEW ResNet 67.0 on ResNet-18/34 baselinesImproved
increase time steps from 4 to 8 with SpikePack (ablation)Spikeformer-8-512 · ImageNet-1ktop-1 accuracy (%): 78.4 -> 80.1Improved
increase time steps from 2 to 8 with SpikePack (ablation)Spikeformer-8-512 · ImageNet-1ktraining memory (MB/img): 150.5 -> 150.5No clear effect
use SpikePack for ANN-to-SNN conversion instead of other conversion methodsResNet-34 / ViT-L/14 · ImageNet-1k conversiontop-1 accuracy (%): 77.92 (ResNet-34), 88.27 (ViT-L/14) at few time stepsImproved
increase time steps for SpikePack on neuromorphic data (ablation)SpikePack · CIFAR10-DVSaccuracy (%): 68.1 -> 84.6Improved
increase time steps from 10 to 16 for SpikePack (ablation)SpikePack · DVS-Gestureaccuracy (%): 97.4 -> 97.4No clear effect
increase time steps from 6 to 10 for SpikePack detector (ablation)SpikePack w/ DINO-r50 · COCO 2017 valmAP@50:95 (%): 48.5 -> 50.1Improved
increase time steps from 6 to 10 for SpikePack segmentation (ablation)SpikePack w/ Segformer-b2 · ADE20KmIoU (%): 42.8 -> 45.6Improved
set membrane time constant to 4.0 instead of 2.0 (ablation)SpikePack ResNet-34 · ImageNet-1kerror rate (%)Improved
replace LIF with SpikePack on FPGA neuromorphic processorResNet-34 · FPGA simulationlatency: 29.1 ms -> 23msImproved
replace LIF with SpikePack on FPGA neuromorphic processorResNet-34 · FPGA simulationenergy: 23.8 mJ -> 18.6mJImproved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy