Object-Anchored Composed Image Retrieval on OACIRR 1.0 (test)
77.15Fashion RID@1AdaFocal (ViT-G)
Evaluation Results
| Method | Links | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| AdaFocal (ViT-G)Domain=OACIR, Pretraining Data=OACIRR (Ours), Backbone=ViT-G2026.04 | 77.15 | 65.31 | 86.88 | 78.42 | 53.63 | 92.22 | 91.86 | 74.11 | 95.39 | 82.92 | 58.47 | 91.63 | 79 | |
| AdaFocal (ViT-L)Domain=OACIR, Pretraining Data=OACIRR (Ours), Backbone=ViT-L2026.04 | 72.6 | 61.95 | 85.3 | 75.68 | 51.87 | 90.04 | 87.76 | 69.94 | 93.32 | 80.5 | 57.55 | 90.25 | 76.4 | |
| SPRC (ViT-G)Domain=CIR, Pretraining Data=OACIRR (Ours), Backbone=ViT-G2026.04 | 65.25 | 58.51 | 80.89 | 72.87 | 49.82 | 89.57 | 86.05 | 70.61 | 93.68 | 76.32 | 56.04 | 89 | 74.05 | |
| SPRC (ViT-L)Domain=CIR, Pretraining Data=OACIRR (Ours), Backbone=ViT-L2026.04 | 61.09 | 54.8 | 75.85 | 68.99 | 46.48 | 86.95 | 80.29 | 67.14 | 90.41 | 72.62 | 54.27 | 86.11 | 70.42 | |
| U-MARVELDomain=UMR, Pretraining Data=M-BEIR + NLI2026.04 | 46.05 | 40.38 | 60.59 | 62.92 | 39.96 | 74.9 | 83.26 | 54.69 | 84.13 | 69.81 | 37.67 | 73.08 | 60.62 | |
| GME (7B)Domain=UMR, Pretraining Data=UMRB [52], Model size=7B2026.04 | 44.98 | 39.24 | 60.18 | 63.11 | 38.34 | 75.38 | 83.44 | 54.6 | 84.15 | 77.11 | 47.09 | 82.69 | 62.53 | |
| MM-EmbedDomain=UMR, Pretraining Data=M-BEIR + MTEB [35]2026.04 | 41.38 | 34.55 | 52.5 | 53.21 | 30.06 | 62.8 | 71.03 | 41.47 | 71.15 | 78.85 | 38.88 | 79.32 | 54.6 | |
| GME (2B)Domain=UMR, Pretraining Data=UMRB [52], Model size=2B2026.04 | 38.13 | 32.14 | 51.5 | 58.84 | 31.6 | 66.03 | 76.89 | 44.11 | 74.2 | 73.86 | 38.99 | 75.61 | 55.16 | |
| SPRC (ViT-G)Domain=CIR, Pretraining Data=CIRR [31], Backbone=ViT-G2026.04 | 28.62 | 25.79 | 44.48 | 25.13 | 15.92 | 37.06 | 54.39 | 34.85 | 62.31 | 40.41 | 26.29 | 52.39 | 37.3 | |
| SPRC (ViT-L)Domain=CIR, Pretraining Data=CIRR [31], Backbone=ViT-L2026.04 | 28.54 | 25.49 | 44.26 | 22.47 | 15.23 | 36.78 | 52.55 | 33.35 | 61.47 | 37.31 | 24.2 | 49.99 | 35.97 | |
| UniIR-BLIPFFDomain=UMR, Pretraining Data=M-BEIR [43]2026.04 | 28.53 | 22.41 | 39.63 | 37.21 | 19.97 | 46.51 | 37.76 | 20.98 | 43.19 | 31.71 | 17.14 | 52.12 | 33.1 | |
| LamRA-RetDomain=UMR, Pretraining Data=M-BEIR + NLI [36]2026.04 | 27.45 | 21.63 | 37.1 | 61.03 | 35.44 | 74.51 | 69.45 | 39.53 | 70.25 | 58.64 | 32.58 | 68.74 | 49.7 | |
| UniIR-CLIPSFDomain=UMR, Pretraining Data=M-BEIR [43]2026.04 | 17.33 | 12.26 | 24.76 | 32.67 | 16.95 | 41.89 | 33.71 | 18.22 | 40.1 | 29.47 | 15.51 | 43.24 | 27.18 | |
| LinCIRDomain=ZS-CIR, Pretraining Data=CC3M [9]2026.04 | 15.78 | 12.04 | 21.82 | 5.55 | 2.23 | 7.28 | 47.55 | 14.63 | 34.91 | 42.76 | 19.57 | 47.15 | 22.61 | |
| Pic2WordDomain=ZS-CIR, Pretraining Data=CC3M [9]2026.04 | 14.98 | 11.15 | 21.55 | 12.07 | 4.07 | 11.32 | 45.95 | 13.66 | 34.19 | 55.98 | 20.99 | 52.12 | 24.84 |