AIM (CLIP ViT-L/14, 32x224)
Reported on 5 benchmarks across 3 tasks · 1 paper · 2 SOTA
Note: results are matched by exact model name. Different papers may use the same name for different model variants.
Computer Vision3 results
- Top-1 Accuracy· uses extra data· 2023-02-0680.4best: 85.9 (InternVideo2-6B)
- Acc@1· uses extra data· 2023-02-0687.5best: 93.6 (OmniVec2)
- Acc@5· uses extra data· 2023-02-0697.7best: 98.9 (TubeViT-H (ImageNet-1k))
Robots1 result
- Accuracy· uses extra data· 2023-02-06SOTA90.6best: 94.9 (LVMAE)
Time Series1 result
- Accuracy· uses extra data· 2023-02-06SOTA90.6best: 94.9 (LVMAE)