Attributing Multimodal Foundation Model Errors on ImageNet misclassified samples (val)
66.39Avg Highest Confidence (0-25%)
LIMA
Updated 14d ago
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
2025.04 | 66.39 | 74.87 | 77.36 | 78.72 | 42.37 | |
2025.04 | 64.4 | 73.07 | 76.76 | 78.78 | 41.51 | |
2025.04 | 56.77 | 70.3 | 74.79 | 75.83 | 47.89 | |
2025.04 | 54.01 | 63.99 | 67.53 | 69.55 | 34.75 | |
2025.04 | 53.51 | 68.16 | 74.45 | 75.57 | 46.81 | |
2025.04 | 44.36 | 58.65 | 61.57 | 63.48 | 39.29 | |
2025.04 | 38.43 | 53.11 | 58.24 | 59.78 | 34.83 | |
2025.04 | 27.1 | 39.85 | 45.58 | 48.79 | 18.27 | |
2025.04 | 24.27 | 40.45 | 47.57 | 51.64 | 22.37 | |
2025.04 | 22.02 | 34.04 | 37.67 | 41.23 | 13.97 | |
2025.04 | 21.48 | 27.23 | 30.51 | 34.62 | 10.62 | |
2025.04 | 20.5 | 29.33 | 33.6 | 35.06 | 18.12 | |
2025.04 | 20.12 | 30.73 | 35.82 | 39.89 | 20.17 | |
2025.04 | 17.11 | 27.26 | 32.6 | 36.36 | 21.8 | |
2025.04 | 14.04 | 28.02 | 37.55 | 42.45 | 17.21 | |
2025.04 | 12.99 | 25.93 | 35.32 | 42.61 | 11.29 | |
2025.04 | 12.68 | 23.67 | 31.37 | 39.02 | 12.36 | |
2025.04 | 10.64 | 19.01 | 26.58 | 34.63 | 11.24 | |
2025.04 | 10.04 | 18.61 | 25.61 | 33.97 | 10.86 | |
2025.04 | 9.67 | 19.02 | 25.3 | 29.2 | 15 | |
2025.04 | 7.87 | 13.8 | 17.79 | 21.12 | 8.61 | |
2025.04 | 4.14 | 11.23 | 16.28 | 23.48 | 8.16 |