Linear Regression In-context Learning on Garg Linear Regression d=10
0.996R-squaredTransformer
Evaluation results
| Method | Links | |
|---|---|---|
| Transformer2026.07 | 0.996 | |
| Spikformer2026.07 | 0.977 | |
| DendriCNN2026.07 | 0.824 | |
| DendriCLTraining Steps=50k2026.07 | 0.815 | |
| DendriCL2026.07 | 0.811 | |
| DendriCLNumber of Layers=22026.07 | 0.811 | |
| Pure LIF2026.07 | 0.801 | |
| Linear Tx2026.07 | 0.77 | |
| MLP-Mixer2026.07 | 0.691 | |
| Active Dendrites2026.07 | 0.668 | |
| GRU2026.07 | 0.666 | |
| LTC2026.07 | 0.409 | |
| LSNN2026.07 | 0.011 | |
| Pure MLP2026.07 | 0.008 | |
| SpikingSSM2026.07 | 0.007 |