Image Classification on LADA-Sculpture (OOD)
82.05AccuracyCLIP ZS
Evaluation results
| Method | Links | |
|---|---|---|
| CLIP ZSBackbone=CLIP ViT-L/14, Concept=none2025.03 | 82.05 | |
| CLIP ZS2025.03 | 82.05 | |
| LaBO + DDOBackbone=CLIP ViT-L/14, Concept=LLM2025.03 | 80 | |
| LaBO‡Concept=LLM, Method=+DDO2025.03 | 80 | |
| PCBM+ + DDOBackbone=CLIP ViT-L/14, Concept=ConceptNet2025.03 | 79.74 | |
| PCBM†Concept=ConceptNet, Method=+DDO, Backbone=CLIP2025.03 | 79.74 | |
| LaBO + DDOBackbone=CLIP ViT-B/32, Concept=LLM2025.03 | 79.69 | |
| LaBO baselineBackbone=CLIP ViT-B/32, Concept=LLM2025.03 | 77.63 | |
| PCBM+ baselineBackbone=CLIP ViT-L/14, Concept=ConceptNet2025.03 | 76.69 | |
| PCBM†Concept=ConceptNet, Method=baseline, Backbone=CLIP2025.03 | 76.69 | |
| CLIP-CBM + DDOBackbone=CLIP ViT-L/14, Concept=human2025.03 | 75.76 | |
| CLIP-CBMConcept=human, Method=+DDO2025.03 | 75.76 | |
| LaBO baselineBackbone=CLIP ViT-L/14, Concept=LLM2025.03 | 74.56 | |
| LaBO‡Concept=LLM, Method=baseline2025.03 | 74.56 | |
| CLIP LPBackbone=CLIP ViT-L/14, Concept=none2025.03 | 74.4 | |
| CLIP LP2025.03 | 74.4 | |
| CLIP-CBM baselineBackbone=CLIP ViT-L/14, Concept=human2025.03 | 70.44 | |
| CLIP-CBMConcept=human, Method=baseline2025.03 | 70.44 | |
| CLIP LPBackbone=CLIP ViT-B/32, Concept=none2025.03 | 65.12 | |
| PCBM+ + DDOBackbone=CLIP ViT-B/32, Concept=ConceptNet2025.03 | 57.96 | |
| PCBM+ baselineBackbone=CLIP ViT-B/32, Concept=ConceptNet2025.03 | 55.64 | |
| CLIP-CBM + DDOBackbone=CLIP ViT-B/32, Concept=human2025.03 | 53.7 | |
| CLIP ZSBackbone=CLIP ViT-B/32, Concept=none2025.03 | 48.98 | |
| CLIP-CBM baselineBackbone=CLIP ViT-B/32, Concept=human2025.03 | 48.89 |