VGG
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Privacy-preserving neural network inference | VGG (inference) | Inference Latency (s)25.6 | 18 | |
| Cell Counting | VGG (test) | MAE4.1 | 14 | |
| Audio classification | VGG-S | Top-1 Accuracy80.1 | 12 | |
| Cell Counting | VGG → ADI | MAE9.2 | 7 | |
| Pose Estimation | VGG multi-view v1 (Wadham College) | Loss0.0001 | 5 | |
| Cell Counting | VGG → DCC | MAE2.7 | 5 | |
| Untargeted Adversarial Attack | VGG-19 | Fooling Rate100 | 5 | |
| RGB-to-X retrieval | VGG-S | R@130.1 | 4 | |
| CNN Inference Acceleration | VGG 16-layer (inference) | Conv Latency (ms)72.5 | 4 | |
| Privacy-Preserving Neural Network Inference (Online Phase) | VGG WAN1 | Latency (s)224 | 3 | |
| Privacy-Preserving Neural Network Inference (Online Phase) | VGG LAN | Communication (MiB)2,410 | 3 | |
| Secure Neural Network Inference | VGG WAN1 | Inference Latency (s)1,092 | 3 | |
| Secure Neural Network Inference | VGG LAN | Communication Overhead (MiB)4,318 | 3 | |
| Encrypted Inference | VGG5 | Memory (GB)0.012 | 2 | |
| Model Inversion Defense (LOMMA-GMI Attack) | VGG-16 | Accuracy81.79 | 2 | |
| Graph partitioning | VGG computation graph | Edge Cut4 | 2 |
Showing 16 of 16 rows