Symmetric Multimodal-to-Multimodal Retrieval on sym-MM2MM 1.0 (test)
77.57R@1SOLAR-C
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| SOLAR-CCategory=Unsupervised, Encoder-based, #Param.(B)=0.20, #Dim.=768, FPS=9222026.05 | 77.57 | 97.2 | 97.66 | 90.81 | 84.58 | 87.69 | |
| SOLAR-B+DCategory=Unsupervised, Encoder-based, #Param.(B)=0.71, #Dim.=768, FPS=5202026.05 | 72.9 | 93.93 | 95.79 | 87.54 | 85.51 | 86.53 | |
| UniMECategory=Supervised, VLM-based, #Param.(B)=7.49, #Dim.=3584, FPS=182026.05 | 59.81 | 93.93 | 95.33 | 83.02 | 73.36 | 78.19 | |
| mmE5Category=Supervised, VLM-based, #Param.(B)=10.12, #Dim.=4096, FPS=42026.05 | 57.94 | 97.66 | 98.13 | 84.58 | 76.64 | 80.61 | |
| Qwen3-VL-EmbeddingCategory=Supervised, VLM-based, #Param.(B)=7.75, #Dim.=4096, FPS=192026.05 | 56.54 | 92.06 | 94.86 | 81.15 | 74.77 | 77.96 | |
| GMECategory=Supervised, VLM-based, #Param.(B)=7.75, #Dim.=3584, FPS=192026.05 | 56.07 | 91.12 | 93.93 | 80.37 | 74.77 | 77.57 | |
| CLIP-SFCategory=Supervised, Encoder-based, #Param.(B)=0.43, #Dim.=768, FPS=3612026.05 | 55.61 | 94.86 | 97.2 | 82.55 | 73.36 | 77.96 | |
| MM-EmbedCategory=Supervised, VLM-based, #Param.(B)=7.75, #Dim.=4096, FPS=192026.05 | 55.61 | 94.39 | 96.26 | 82.09 | 75.7 | 78.89 | |
| LamRACategory=Supervised, VLM-based, #Param.(B)=7.75, #Dim.=3584, FPS=192026.05 | 53.74 | 82.71 | 90.65 | 75.7 | 78.5 | 77.1 | |
| CLIP-SF-ZSCategory=Unsupervised, Encoder-based, #Param.(B)=0.15, #Dim.=512, FPS=11572026.05 | 53.27 | 92.99 | 94.39 | 80.22 | 71.03 | 75.62 | |
| UniteCategory=Supervised, VLM-based, #Param.(B)=7.75, #Dim.=3584, FPS=192026.05 | 52.8 | 91.59 | 95.33 | 79.91 | 75.23 | 77.57 | |
| VLM2VecCategory=Supervised, VLM-based, #Param.(B)=7.75, #Dim.=3584, FPS=192026.05 | 52.33 | 79.91 | 85.05 | 72.43 | 71.5 | 71.96 | |
| BGE-VLCategory=Supervised, Encoder-based, #Param.(B)=0.15, #Dim.=512, FPS=11572026.05 | 29.91 | 57.94 | 61.68 | 49.84 | 64.02 | 56.93 |