Image Classification on LADV-3D (ID)
99.93AccuracyLaBO + DDO
Evaluation results
| Method | Links | |
|---|---|---|
| LaBO + DDOBackbone=CLIP ViT-L/14, Concept=LLM2025.03 | 99.93 | |
| LaBO‡Concept=LLM, Method=+DDO2025.03 | 99.93 | |
| LaBO baselineBackbone=CLIP ViT-L/14, Concept=LLM2025.03 | 99.9 | |
| LaBO‡Concept=LLM, Method=baseline2025.03 | 99.9 | |
| LaBO baselineBackbone=CLIP ViT-B/32, Concept=LLM2025.03 | 99.44 | |
| LaBO + DDOBackbone=CLIP ViT-B/32, Concept=LLM2025.03 | 99.13 | |
| PCBM+ + DDOBackbone=CLIP ViT-L/14, Concept=ConceptNet2025.03 | 94.82 | |
| PCBM†Concept=ConceptNet, Method=+DDO, Backbone=CLIP2025.03 | 94.82 | |
| PCBM+ baselineBackbone=CLIP ViT-L/14, Concept=ConceptNet2025.03 | 94.71 | |
| PCBM†Concept=ConceptNet, Method=baseline, Backbone=CLIP2025.03 | 94.71 | |
| CLIP LPBackbone=CLIP ViT-L/14, Concept=none2025.03 | 93.68 | |
| CLIP LP2025.03 | 93.68 | |
| CLIP-CBM + DDOBackbone=CLIP ViT-L/14, Concept=human2025.03 | 92.59 | |
| CLIP-CBMConcept=human, Method=+DDO2025.03 | 92.59 | |
| CLIP-CBM baselineBackbone=CLIP ViT-L/14, Concept=human2025.03 | 92.21 | |
| CLIP-CBMConcept=human, Method=baseline2025.03 | 92.21 | |
| PCBM+ baselineBackbone=CLIP ViT-B/32, Concept=ConceptNet2025.03 | 90.11 | |
| PCBM+ + DDOBackbone=CLIP ViT-B/32, Concept=ConceptNet2025.03 | 90.03 | |
| CLIP-CBM baselineBackbone=CLIP ViT-B/32, Concept=human2025.03 | 87.85 | |
| CLIP-CBM + DDOBackbone=CLIP ViT-B/32, Concept=human2025.03 | 87.29 | |
| CLIP ZSBackbone=CLIP ViT-B/32, Concept=none2025.03 | 84.36 | |
| CLIP ZSBackbone=CLIP ViT-L/14, Concept=none2025.03 | 71.82 | |
| CLIP ZS2025.03 | 71.82 | |
| CLIP LPBackbone=CLIP ViT-B/32, Concept=none2025.03 | 68.28 |