Image Classification on CIFAR-10 v1 (test)
98.62AccuracyPMSM
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| PMSMPolarity=true, Multi-Spike=true, Direct Training=false2025.08 | 98.62 | — | — | 98.85 | 1 | |
| SpikeZIP-TFPolarity=true, Multi-Spike=false, Direct Training=false2025.08 | 97.7 | — | — | 99.2 | 16 | |
| SpikedAttentionPolarity=false, Multi-Spike=false, Direct Training=false2025.08 | 97.3 | — | — | 97.5 | 24 | |
| MSTPolarity=false, Multi-Spike=false, Direct Training=true2025.08 | 97.27 | — | — | 98.14 | 256 | |
| SdformerPolarity=false, Multi-Spike=false, Direct Training=true2025.08 | 95.6 | — | — | — | 4 | |
| SpikFormerPolarity=false, Multi-Spike=false, Direct Training=true2025.08 | 95.51 | — | — | — | 4 | |
| LaDP-FLepsilon=0.5, Backbone=ResNet-182026.01 | 90.06 | — | — | — | — | |
| AdapLDPepsilon=0.5, Backbone=ResNet-182026.01 | 88.39 | 1.89 | 4.63 | — | — | |
| LaDP-FLepsilon=0.4, Backbone=ResNet-182026.01 | 87.12 | — | — | — | — | |
| DM (beta=0)noise ratio (r)=0, beta=0, backbone=GoogLeNet V12019.05 | 86 | — | — | — | — | |
| DM (lambda=0.0)noise ratio (r)=0, lambda=0, backbone=GoogLeNet V12019.05 | 86 | — | — | — | — | |
| DM (lambda=0.5)noise ratio (r)=0, lambda=0.5, backbone=GoogLeNet V12019.05 | 86 | — | — | — | — | |
| CCEnoise ratio (r)=0, backbone=GoogLeNet V12019.05 | 85 | — | — | — | — | |
| MSE-DNnoise ratio (r)=0, backbone=GoogLeNet V12019.05 | 85 | — | — | — | — | |
| Sensitive DPepsilon=0.5, Backbone=ResNet-182026.01 | 84.68 | 6.35 | 18.46 | — | — | |
| CCE-DNnoise ratio (r)=0, backbone=GoogLeNet V12019.05 | 84 | — | — | — | — | |
| AdapLDPepsilon=0.4, Backbone=ResNet-182026.01 | 83.28 | 4.61 | — | — | — | |
| GCEnoise ratio (r)=0, backbone=GoogLeNet V12019.05 | 83 | — | — | — | — | |
| DM (lambda=0.5)noise ratio (r)=0.2, lambda=0.5, backbone=GoogLeNet V12019.05 | 83 | — | — | — | — | |
| LaDP-FLepsilon=0.3, Backbone=ResNet-182026.01 | 82.29 | — | — | — | — | |
| GCE-DNnoise ratio (r)=0.2, backbone=GoogLeNet V12019.05 | 82 | — | — | — | — | |
| Mosaicω=1.0, Heterogeneity=Standard, Backbone=ResNet-182025.05 | 81.95 | — | — | — | — | |
| pFedFDAω=1.0, Heterogeneity=Standard, Backbone=ResNet-182025.05 | 81.73 | — | — | — | — | |
| PA3Fedω=1.0, Heterogeneity=Standard, Backbone=ResNet-182025.05 | 81.56 | — | — | — | — | |
| FedOptω=1.0, Heterogeneity=Standard, Backbone=ResNet-182025.05 | 81.3 | — | — | — | — | |
| FedInitω=1.0, Heterogeneity=Standard, Backbone=ResNet-182025.05 | 81.18 | — | — | — | — | |
| GCEnoise ratio (r)=0.2, backbone=GoogLeNet V12019.05 | 81 | — | — | — | — | |
| DFRDω=1.0, Heterogeneity=Standard, Backbone=ResNet-182025.05 | 80.27 | — | — | — | — | |
| Self-pacednoise ratio (r)=0, backbone=GoogLeNet V1, source=MentorNet (Jiang et al., 2018)2019.05 | 80 | — | — | — | — | |
| MSEnoise ratio (r)=0, backbone=GoogLeNet V12019.05 | 80 | — | — | — | — | |
| DM (lambda=0.5)noise ratio (r)=0.4, lambda=0.5, backbone=GoogLeNet V12019.05 | 80 | — | — | — | — | |
| FedKFDω=1.0, Heterogeneity=Standard, Backbone=ResNet-182025.05 | 79.86 | — | — | — | — | |
| FedAFω=1.0, Heterogeneity=Standard, Backbone=ResNet-182025.05 | 79.81 | — | — | — | — | |
| DENSEω=1.0, Heterogeneity=Standard, Backbone=ResNet-182025.05 | 79.56 | — | — | — | — | |
| DPAepsilon=0.5, Backbone=ResNet-182026.01 | 79.36 | 13.48 | 13.44 | — | — | |
| FedFTGω=1.0, Heterogeneity=Standard, Backbone=ResNet-182025.05 | 79.01 | — | — | — | — | |
| MentorNet PDnoise ratio (r)=0, backbone=GoogLeNet V1, source=MentorNet (Jiang et al., 2018)2019.05 | 79 | — | — | — | — | |
| MentorNet DDnoise ratio (r)=0, backbone=GoogLeNet V1, source=MentorNet (Jiang et al., 2018)2019.05 | 79 | — | — | — | — | |
| GCE-DNnoise ratio (r)=0.4, backbone=GoogLeNet V12019.05 | 79 | — | — | — | — | |
| FedRSω=1.0, Heterogeneity=Standard, Backbone=ResNet-182025.05 | 78.58 | — | — | — | — | |
| Time-Varying DPepsilon=0.5, Backbone=ResNet-182026.01 | 78.51 | 14.71 | 24.75 | — | — | |
| FedAvgω=1.0, Heterogeneity=Standard, Backbone=ResNet-182025.05 | 78.34 | — | — | — | — | |
| Boot-softnoise ratio (r)=0, backbone=GoogLeNet V1, source=MentorNet (Jiang et al., 2018)2019.05 | 78 | — | — | — | — | |
| MSEnoise ratio (r)=0.2, backbone=GoogLeNet V12019.05 | 78 | — | — | — | — | |
| DPAepsilon=0.4, Backbone=ResNet-182026.01 | 77.94 | 11.78 | — | — | — | |
| LaDP-FLepsilon=0.2, Backbone=ResNet-182026.01 | 77.82 | — | — | — | — | |
| AdapLDPepsilon=0.3, Backbone=ResNet-182026.01 | 77.12 | 6.7 | — | — | — | |
| Focal Lossnoise ratio (r)=0, backbone=GoogLeNet V1, source=MentorNet (Jiang et al., 2018)2019.05 | 77 | — | — | — | — | |
| GCEnoise ratio (r)=0.4, backbone=GoogLeNet V12019.05 | 77 | — | — | — | — | |
| GCE-DNnoise ratio (r)=0, backbone=GoogLeNet V12019.05 | 77 | — | — | — | — | |
| DM (lambda=0.0)noise ratio (r)=0.2, lambda=0, backbone=GoogLeNet V12019.05 | 77 | — | — | — | — | |
| AdapLDPepsilon=0.5, Backbone=CNN2026.01 | 76.49 | -1.48 | 2.55 | — | — | |
| Mosaicω=1.0, Heterogeneity=Stronger, Backbone=ResNet-182025.05 | 76.27 | — | — | — | — | |
| Forgettingnoise ratio (r)=0, backbone=GoogLeNet V1, source=MentorNet (Jiang et al., 2018)2019.05 | 76 | — | — | — | — | |
| MentorNet DDnoise ratio (r)=0.2, backbone=GoogLeNet V1, source=MentorNet (Jiang et al., 2018)2019.05 | 76 | — | — | — | — | |
| CCE-DNnoise ratio (r)=0.4, backbone=GoogLeNet V12019.05 | 76 | — | — | — | — | |
| MSE-DNnoise ratio (r)=0.4, backbone=GoogLeNet V12019.05 | 76 | — | — | — | — | |
| LaDP-FLepsilon=0.5, Backbone=CNN2026.01 | 75.36 | — | — | — | — | |
| CCE-DNnoise ratio (r)=0.2, backbone=GoogLeNet V12019.05 | 75 | — | — | — | — | |
| DM (lambda=0.0)noise ratio (r)=0.4, lambda=0, backbone=GoogLeNet V12019.05 | 75 | — | — | — | — | |
| Self-pacednoise ratio (r)=0.2, backbone=GoogLeNet V1, source=MentorNet (Jiang et al., 2018)2019.05 | 74 | — | — | — | — | |
| Focal Lossnoise ratio (r)=0.2, backbone=GoogLeNet V1, source=MentorNet (Jiang et al., 2018)2019.05 | 74 | — | — | — | — | |
| MentorNet PDnoise ratio (r)=0.2, backbone=GoogLeNet V1, source=MentorNet (Jiang et al., 2018)2019.05 | 74 | — | — | — | — | |
| CCEnoise ratio (r)=0.2, backbone=GoogLeNet V12019.05 | 74 | — | — | — | — | |
| CCEnoise ratio (r)=0.4, backbone=GoogLeNet V12019.05 | 74 | — | — | — | — | |
| AdapLDPepsilon=0.2, Backbone=ResNet-182026.01 | 73.9 | 5.3 | — | — | — | |
| DPAepsilon=0.3, Backbone=ResNet-182026.01 | 73.77 | 11.55 | — | — | — | |
| Boot-softnoise ratio (r)=0.2, backbone=GoogLeNet V1, source=MentorNet (Jiang et al., 2018)2019.05 | 73 | — | — | — | — | |
| MSEnoise ratio (r)=0.4, backbone=GoogLeNet V12019.05 | 73 | — | — | — | — | |
| Sensitive DPepsilon=0.5, Backbone=CNN2026.01 | 71.07 | 6.04 | 129.62 | — | — | |
| Forgettingnoise ratio (r)=0.2, backbone=GoogLeNet V1, source=MentorNet (Jiang et al., 2018)2019.05 | 71 | — | — | — | — | |
| DPAepsilon=0.5, Backbone=CNN2026.01 | 70.66 | 6.65 | 25.28 | — | — | |
| Full DPepsilon=0.5, Backbone=ResNet-182026.01 | 70.59 | 27.58 | 36.23 | — | — | |
| Time-Varying DPepsilon=0.5, Backbone=CNN2026.01 | 69.74 | 8.06 | 87.75 | — | — | |
| Sensitive DPepsilon=0.4, Backbone=ResNet-182026.01 | 69.68 | 25.03 | — | — | — | |
| Sensitive DPepsilon=0.3, Backbone=ResNet-182026.01 | 67.64 | 21.66 | — | — | — | |
| Time-Varying DPepsilon=0.4, Backbone=ResNet-182026.01 | 66.79 | 30.44 | — | — | — | |
| DPAepsilon=0.2, Backbone=ResNet-182026.01 | 66.54 | 16.95 | — | — | — | |
| DM (beta=0)noise ratio (r)=0.2, beta=0, backbone=GoogLeNet V12019.05 | 66 | — | — | — | — | |
| pFedFDAω=1.0, Heterogeneity=Stronger, Backbone=ResNet-182025.05 | 64.92 | — | — | — | — | |
| Sensitive DPepsilon=0.2, Backbone=ResNet-182026.01 | 64.43 | 20.78 | — | — | — | |
| Time-Varying DPepsilon=0.2, Backbone=ResNet-182026.01 | 63.65 | 22.26 | — | — | — | |
| DENSEω=1.0, Heterogeneity=Stronger, Backbone=ResNet-182025.05 | 63.61 | — | — | — | — | |
| FedRSω=1.0, Heterogeneity=Stronger, Backbone=ResNet-182025.05 | 63.37 | — | — | — | — | |
| Full DPepsilon=0.5, Backbone=CNN2026.01 | 63.19 | 19.26 | 94.96 | — | — | |
| FedKFDω=1.0, Heterogeneity=Stronger, Backbone=ResNet-182025.05 | 63.08 | — | — | — | — | |
| PA3Fedω=1.0, Heterogeneity=Stronger, Backbone=ResNet-182025.05 | 63.02 | — | — | — | — | |
| DFRDω=1.0, Heterogeneity=Stronger, Backbone=ResNet-182025.05 | 63 | — | — | — | — | |
| FedInitω=1.0, Heterogeneity=Stronger, Backbone=ResNet-182025.05 | 62.88 | — | — | — | — | |
| Time-Varying DPepsilon=0.3, Backbone=ResNet-182026.01 | 62.54 | 31.58 | — | — | — | |
| Mosaicω=0.1, Heterogeneity=Standard, Backbone=ResNet-182025.05 | 62.54 | — | — | — | — | |
| FedOptω=1.0, Heterogeneity=Stronger, Backbone=ResNet-182025.05 | 62.45 | — | — | — | — | |
| FedFTGω=1.0, Heterogeneity=Stronger, Backbone=ResNet-182025.05 | 61.96 | — | — | — | — | |
| PA3Fedω=0.1, Heterogeneity=Standard, Backbone=ResNet-182025.05 | 61.5 | — | — | — | — | |
| FedAFω=1.0, Heterogeneity=Stronger, Backbone=ResNet-182025.05 | 61.31 | — | — | — | — | |
| FedAvgω=1.0, Heterogeneity=Stronger, Backbone=ResNet-182025.05 | 61.23 | — | — | — | — | |
| Full DPepsilon=0.4, Backbone=ResNet-182026.01 | 60.71 | 43.5 | — | — | — | |
| pFedFDAω=0.1, Heterogeneity=Standard, Backbone=ResNet-182025.05 | 60.12 | — | — | — | — | |
| DFRDω=0.1, Heterogeneity=Standard, Backbone=ResNet-182025.05 | 59.8 | — | — | — | — | |
| FedInitω=0.1, Heterogeneity=Standard, Backbone=ResNet-182025.05 | 59.78 | — | — | — | — |