AI Paper Atlas

Research map / Architectures and efficiency / Spiking networks

SMixer: Rethinking Efficient-Training and Event-Driven SNNs

Yijie Lu, Xinhao Luo, Yixing Zhang, Zhiyan Wang, Wentao Li, YanhanWang, Zhi Liu, Zhaokun Zhou, Guoqi Li

ICLR 2026 · Architectures and efficiency › Spiking networks · approach: Training deep spiking networks

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
add DSTSP spatial-temporal pruningSTMixer-8-768 · ImageNet-1Kaccuracy / energy: acc 76.68 -> 76.87; energy 4.45 -> 2.36 mJImproved
add DSTSP to Spikformer-V2 with Spiking-token Mixer (T=4)SpikformerV2->M · ImageNet-1Ktop-1 accuracy: 79.12 -> 78.03Hurt
add DSTSP to SDT-V3 with Spiking-token MixerSDT-V3->M · ImageNet-1Ktop-1 accuracy: 79.25 -> 77.75Hurt
replace spiking self-attention with Spiking-token MixerSpiking Transformers · ImageNet-1Ktop-1 accuracy: 79.05 -> 79.16 (Spikformer V2-8-512)No clear effect
add DSTSP (spatial 0.3, one time step)SMixer · CIFAR10/CIFAR100top-1 accuracy: 96.01 -> 95.67; 81.87 -> 81.03Hurt
add DSTSPSMixer · CIFAR10-DVS and DVS128top-1 accuracy: 83.02 -> 82.34; 98.61 -> 98.26Hurt
add DSTSPSMixer (SpikeYOLO) · COCO val2017mAP@50 and power: mAP 58.9 -> 57.4; power 36.8 -> 21.2 mJMixed
add DSTSPSMixer · time-series forecasting (Metr-la, Pems-bay, Solar, Electricity)R2 / RSENo clear effect
prune low-SIV tokens instead of random or high-SIV tokens (ablation)DSTSP · CIFAR100top-1 accuracy: 81.03 vs 78.11 (high-SIV pruning)Improved
apply DSTSP in both training and inference instead of inference only (ablation)DSTSP · CIFAR100top-1 accuracy: 79.74 -> 81.03Improved
apply DSTSP in training only instead of training and inference (ablation)DSTSP · CIFAR100top-1 accuracy: 81.03 -> 80.95No clear effect
increase spatial pruning ratio from 0.4 to 0.6 (ablation)DSTSP · CIFAR100top-1 accuracy: 80.81 -> 79.13Hurt

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