Multimodal Retrieval on Grocery MAG benchmark
88.8R@1GOMA
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| GOMAMethod Category=Our Method, Feature Backbone=Qwen2-VL2026.05 | 88.8 | 99.7 | 100 | |
| DGFMethod Category=Multimodal Graph Retrieval & Recent MAG Models, Feature Backbone=Qwen2-VL2026.05 | 85.8 | 98.7 | 99.6 | |
| DMGCMethod Category=Multimodal Graph Retrieval & Recent MAG Models, Feature Backbone=Qwen2-VL2026.05 | 56.3 | 74.7 | 80.1 | |
| SmartCLIPMethod Category=Pairwise Image-Text Retrieval Baselines, Feature Backbone=Qwen2-VL2026.05 | 45.8 | 74.1 | 82.3 | |
| CLIP-RefineMethod Category=Pairwise Image-Text Retrieval Baselines, Feature Backbone=Qwen2-VL2026.05 | 38 | 63.7 | 72.2 | |
| VSE++Method Category=Pairwise Image-Text Retrieval Baselines, Feature Backbone=Qwen2-VL2026.05 | 21.6 | 48.6 | 60.8 | |
| LGMRecMethod Category=Multimodal Graph Retrieval & Recent MAG Models, Feature Backbone=Qwen2-VL2026.05 | 19.2 | 47.1 | 60.8 | |
| D2S-VSEMethod Category=Pairwise Image-Text Retrieval Baselines, Feature Backbone=Qwen2-VL2026.05 | 16.5 | 40.8 | 52.5 | |
| LinearMethod Category=Direct Retrieval Baselines, Feature Backbone=Qwen2-VL2026.05 | 13.6 | 33.5 | 45 | |
| MMGCNMethod Category=Multimodal Graph Retrieval & Recent MAG Models, Feature Backbone=Qwen2-VL2026.05 | 10.3 | 31.7 | 45.9 | |
| MGATMethod Category=Multimodal Graph Retrieval & Recent MAG Models, Feature Backbone=Qwen2-VL2026.05 | 6.1 | 20.6 | 31.5 | |
| RawMethod Category=Direct Retrieval Baselines, Feature Backbone=Qwen2-VL2026.05 | 3.5 | 8.3 | 11.3 | |
| GSMNMethod Category=Multimodal Graph Retrieval & Recent MAG Models, Feature Backbone=Qwen2-VL2026.05 | 2.6 | 10.3 | 17.4 |