Binary Classification on NICO++ Car vs. Truck
86.1Average AccuracyBAP
Evaluation results
| Method | Links | ||
|---|---|---|---|
| BAPBackbone=SigLIP 2 ViT-B/16, Evaluation Protocol=Zero-Shot2026.05 | 86.1 | 69.4 | |
| Native BackboneBackbone=SigLIP 2 ViT-B/16, Evaluation Protocol=Zero-Shot2026.05 | 85.2 | 59.4 | |
| BAPBackbone=SigLIP 2 ViT-B/16, Evaluation Protocol=Linear Probing2026.05 | 84.7 | 65.2 | |
| BAP (LP)Backbone=CLIP ConvNeXt-W, Evaluation Protocol=Linear Probing2026.05 | 83.8 | 72.9 | |
| DIALBackbone=SigLIP 2 ViT-B/16, Evaluation Protocol=Zero-Shot2026.05 | 83.5 | 64.1 | |
| Data-Matched ControlBackbone=SigLIP 2 ViT-B/16, Evaluation Protocol=Trained Evaluation2026.05 | 83.4 | 57.1 | |
| Data-Matched ControlBackbone=CLIP ConvNeXt-W, Evaluation Protocol=Linear Probing2026.05 | 81.8 | 59.3 | |
| Native BackboneBackbone=SigLIP 2 ViT-B/16, Evaluation Protocol=Linear Probing2026.05 | 80.5 | 31.2 | |
| Native Backbone (LP)Backbone=CLIP ConvNeXt-W, Evaluation Protocol=Linear Probing2026.05 | 79.6 | 40 | |
| Native Backbone (ZS)Backbone=CLIP ConvNeXt-W, Evaluation Protocol=Zero-shot2026.05 | 79 | 56.5 | |
| BAP (ZS)Backbone=CLIP ConvNeXt-W, Evaluation Protocol=Zero-shot2026.05 | 79 | 63.1 | |
| DIALBackbone=CLIP ConvNeXt-W, Evaluation Protocol=Zero-shot2026.05 | 76.2 | 58.3 |