Multimodal Knowledge Editing on MMQAKE Original Image
53.75M-AccHybrid-DMKG
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Hybrid-DMKGBackbone=LLaVA (7B)2025.11 | 53.75 | 29.9 | |
| Hybrid-DMKGBackbone=BLIP-2 (3.8B)2025.11 | 47.55 | 28.88 | |
| IKEBackbone=LLaVA (7B)2025.11 | 38.93 | 16.38 | |
| Hybrid-DMKGBackbone=MiniGPT-4 (7.8B)2025.11 | 35.86 | 24.73 | |
| IKEBackbone=BLIP-2 (3.8B)2025.11 | 16.64 | 6.16 | |
| IKEBackbone=MiniGPT-4 (7.8B)2025.11 | 15.48 | 6.14 | |
| SERACBackbone=LLaVA (7B)2025.11 | 6.58 | 0 | |
| SERACBackbone=BLIP-2 (3.8B)2025.11 | 5.75 | 0 | |
| FT(QFor)Backbone=MiniGPT-4 (7.8B)2025.11 | 4.69 | 0.44 | |
| FT(QFor)Backbone=LLaVA (7B)2025.11 | 4.63 | 0.44 | |
| FT(QFor)Backbone=BLIP-2 (3.8B)2025.11 | 3.73 | 0.2 | |
| FT(All)Backbone=LLaVA (7B)2025.11 | 1.66 | 0 | |
| MENDBackbone=LLaVA (7B)2025.11 | 0.7 | 0 | |
| FT(All)Backbone=BLIP-2 (3.8B)2025.11 | 0.32 | 0.02 | |
| SERACBackbone=MiniGPT-4 (7.8B)2025.11 | 0.27 | 0 | |
| FT(All)Backbone=MiniGPT-4 (7.8B)2025.11 | 0.08 | 0 | |
| MENDBackbone=MiniGPT-4 (7.8B)2025.11 | 0.07 | 0 | |
| MENDBackbone=BLIP-2 (3.8B)2025.11 | 0.04 | 0 |