Recipe-to-Image Retrieval on Recipe1M 10k setup (test)
61.5R@1SIMMER (V1-7B)
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| SIMMER (V1-7B)Venue=-2026.04 | 61.5 | 1 | 85 | 91 | |
| SIMMERModel Version=V1, Parameter Count=7B2026.04 | 61.5 | 1 | 85 | 91 | |
| TNLBTVenue=MMM’232026.04 | 55.9 | 1 | 80.1 | 86.8 | |
| TNLBT2026.04 | 55.9 | 1 | 80.1 | 86.8 | |
| SIMMER (V1-2B)Venue=-2026.04 | 55.8 | 1 | 81.1 | 88 | |
| SIMMERModel Version=V1, Parameter Count=2B2026.04 | 55.8 | 1 | 81.1 | 88 | |
| Yang et al.Venue=ICMRW’242026.04 | 55.7 | 1 | 80.2 | 87.1 | |
| SIMMER (V2)Venue=-2026.04 | 55.7 | 1 | 81 | 88 | |
| Yang et al.2026.04 | 55.7 | 1 | 80.2 | 87.1 | |
| SIMMERModel Version=V22026.04 | 55.7 | 1 | 81 | 88 | |
| Wang et al.Venue=MM’252026.04 | 52.2 | 1 | 78.4 | 86 | |
| Wang et al.Venue=MM'252026.04 | 52.2 | 1 | 78.4 | 86 | |
| EOMA2026.04 | 50.1 | 1.3 | 76.8 | 84.5 | |
| FMIVenue=Scientific Reports’252026.04 | 49.5 | 1 | 79.2 | 83.1 | |
| FMI2026.04 | 49.5 | 1 | 79.2 | 83.1 | |
| VLPCookBackbone=ViT-B/16, Pre-training Data=Recipe1M+2022.12 | 47.8 | — | 74.1 | 81.8 | |
| VLPCook (R1M+)Training Dataset=Recipe1M+2022.12 | 47.8 | 2 | 74.1 | 81.8 | |
| DARVenue=ECCV’242026.04 | 47.4 | 2 | 75.5 | 84.1 | |
| DAR2026.04 | 47.4 | 2 | 75.5 | 84.1 | |
| VLPCookBackbone=ViT-B/162022.12 | 46.4 | — | 73.1 | 80.9 | |
| VLPCook2022.12 | 46.4 | 2 | 73.1 | 80.9 | |
| VLPCookVenue=CVIU’242026.04 | 46.4 | 2 | 73.1 | 80.9 | |
| VLPCook2026.04 | 46.4 | 2 | 73.1 | 80.9 | |
| CIPVenue=MM’232026.04 | 45.2 | 2 | 73 | 81.8 | |
| CIP2026.04 | 45.2 | 2 | 73 | 81.8 | |
| CREAMYVenue=IEEE Access’242026.04 | 45 | 2 | 71.4 | 80 | |
| CREAMY2026.04 | 45 | 2 | 71.4 | 80 | |
| T-Foodbackbone=CLIP-ViT2022.04 | 44.6 | 2 | 71.2 | 79.7 | |
| T-Food (CLIP-ViT)Backbone=CLIP-ViT2022.12 | 44.6 | — | 71.2 | 79.7 | |
| T-Food (CLIP-ViT)Backbone=CLIP-ViT2022.12 | 44.6 | 2 | 71.2 | 79.7 | |
| T-FoodVenue=CVPRW’222026.04 | 44.6 | 2 | 71.2 | 79.7 | |
| T-Food2026.04 | 44.6 | 2 | 71.2 | 79.7 | |
| FARMVenue=WACV’242026.04 | 44.3 | 2 | 71.5 | 80 | |
| FARM2026.04 | 44.3 | 2 | 71.5 | 80 | |
| MALM2026.04 | 44.2 | 2 | 71.7 | 80.1 | |
| DCA-Food2026.04 | 42.9 | 2 | 69.7 | 78.5 | |
| T-Foodbackbone=ViT2022.04 | 41 | 2 | 67.3 | 75.9 | |
| T-Food (ViT)Backbone=ViT2022.12 | 41 | — | 67.3 | 75.9 | |
| T-Food (ViT)Backbone=ViT2022.12 | 41 | 2 | 67.3 | 75.9 | |
| MMACMR2026.04 | 38.3 | 2.2 | 65.6 | 75.6 | |
| RDE-GANVenue=MM’212026.04 | 38.2 | 3 | 57.7 | 65.8 | |
| RDE-GAN2026.04 | 38.2 | 3 | 57.7 | 65.8 | |
| UTE-FCLVenue=Knowledge-Based Systems’242026.04 | 36.5 | 3 | 64.7 | 74.8 | |
| UTE-FCL2026.04 | 36.5 | 3 | 64.7 | 74.8 | |
| LMF-CSF2026.04 | 34.3 | 3 | 62.5 | 72.8 | |
| H-Tbackbone=ViT2022.04 | 33.7 | 3 | 62.2 | 72.7 | |
| H-T (ViT)Backbone=ViT2022.12 | 33.7 | — | 62.2 | 72.7 | |
| H-T (ViT)Backbone=ViT2022.12 | 33.7 | 3 | 62.2 | 72.7 | |
| MFRF2026.04 | 33.7 | 4.4 | 59.3 | 68.5 | |
| X-MRS2022.04 | 33 | 3 | 60.4 | 70.7 | |
| X-MRS2022.12 | 33 | — | 60.4 | 70.7 | |
| X-MRS2022.12 | 33 | 3 | 60.4 | 70.7 | |
| X-MRSVenue=MM’212026.04 | 33 | 3 | 60.4 | 70.7 | |
| X-MRS2026.04 | 33 | 3 | 60.4 | 70.7 | |
| M-SIA2022.04 | 30.3 | 4 | 55.6 | 66.5 | |
| M-SIA2022.12 | 30.3 | — | 55.6 | 66.5 | |
| M-SIA2022.12 | 30.3 | 4 | 55.6 | 66.5 | |
| M-SIA2026.04 | 30.3 | 4 | 55.6 | 66.5 | |
| Wang et al.Venue=MM'222026.04 | 29.3 | — | 57.3 | 69 | |
| LCWF-GI2026.04 | 28.6 | 4 | 55.8 | 67.5 | |
| H-T2022.04 | 28.3 | 4 | 56.5 | 68.1 | |
| H-TVenue=CVPR’212026.04 | 28.3 | 4 | 56.5 | 68.1 | |
| H-T2026.04 | 28.3 | 4 | 56.5 | 68.1 | |
| SEJE2022.04 | 27.2 | 4 | 54.4 | 66.1 | |
| SEJE2022.12 | 27.2 | — | 54.4 | 66.1 | |
| SEJE2022.12 | 27.2 | 4 | 54.4 | 66.1 | |
| JEMA2026.04 | 27.2 | 4 | 54.4 | 66.1 | |
| SEJE2026.04 | 27.2 | 4 | 54.4 | 66.1 | |
| MSJE2022.04 | 26.2 | 5 | 52.5 | 64.1 | |
| MSJE2022.12 | 26.2 | — | 52.5 | 64.1 | |
| MSJE2022.12 | 26.2 | 5 | 52.5 | 64.1 | |
| MSJE2026.04 | 26.2 | 5 | 52.5 | 64.1 | |
| HF-ICMA2022.04 | 25.6 | 4.2 | 54.8 | 67.3 | |
| HF-ICMA2022.12 | 25.6 | — | 54.8 | 67.3 | |
| HF-ICMA2022.12 | 25.6 | 4.2 | 54.8 | 67.3 | |
| HF-ICMA2026.04 | 25.6 | 4.2 | 54.8 | 67.3 | |
| SCAN2022.04 | 25.3 | 5.1 | 50.6 | 61.6 | |
| SCAN2022.12 | 25.3 | — | 50.6 | 61.6 | |
| SCAN2022.12 | 25.3 | 5.1 | 50.6 | 61.6 | |
| SCAN2020.03 | 25.3 | 5.1 | 50.6 | 61.6 | |
| SCAN2026.04 | 25.3 | 5.1 | 50.6 | 61.6 | |
| IMHF2022.04 | 24.9 | 5.8 | 48.3 | 59.4 | |
| IMHF2022.12 | 24.9 | — | 48.3 | 59.4 | |
| IMHF2022.12 | 24.9 | 5.8 | 48.3 | 59.4 | |
| IMHF2026.04 | 24.9 | 5.8 | 48.3 | 59.4 | |
| Wang et al.2022.04 | 24.6 | 5.6 | 50 | 61 | |
| Wang et. al2022.12 | 24.6 | — | 50 | 61 | |
| Wang et al.2022.12 | 24.6 | 5.6 | 50 | 61 | |
| Wang et al.Venue=IEEE TPAMI'232026.04 | 24.6 | 5.6 | 50 | 61 | |
| ACME2022.04 | 24.4 | 6 | 47.9 | 59 | |
| ACME2022.12 | 24.4 | — | 47.9 | 59 | |
| ACME2022.12 | 24.4 | 6 | 47.9 | 59 | |
| ACME2020.03 | 24.4 | 6 | 47.9 | 59 | |
| ACMEVenue=CVPR’192026.04 | 24.4 | 6 | 47.9 | 59 | |
| ACME2026.04 | 24.4 | 6 | 47.9 | 59 | |
| SN2022.04 | 23.4 | 7 | 47.3 | 57.9 | |
| SN2022.12 | 23.4 | — | 47.3 | 57.9 | |
| SN2022.12 | 23.4 | 7 | 47.3 | 57.9 | |
| tri-pro2020.03 | 23.4 | 7 | 47.3 | 57.9 | |
| SN2026.04 | 23.4 | 7 | 47.3 | 57.9 |