Image-to-Recipe Retrieval on Recipe1M 1k setup (test)
87.5Recall@1SIMMER
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| SIMMERVersion=V1-7B2026.04 | 87.5 | 1 | 98 | 99.2 | |
| SIMMER (V1-7B)Venue=-2026.04 | 87.5 | 1 | 98 | 99.2 | |
| SIMMERVersion=V1-2B2026.04 | 84.1 | 1 | 97.3 | 98.8 | |
| SIMMER (V1-2B)Venue=-2026.04 | 84.1 | 1 | 97.3 | 98.8 | |
| SIMMERVersion=V22026.04 | 83.8 | 1 | 96.9 | 98.7 | |
| SIMMER (V2)Venue=-2026.04 | 83.8 | 1 | 96.9 | 98.7 | |
| Yang et al.Venue=ICMRW’242026.04 | 81.8 | 1 | 95.9 | 97.8 | |
| Yang et al.Venue=ICMRW’242026.04 | 81.8 | 1 | 95.9 | 97.8 | |
| TNLBTVenue=MMM’232026.04 | 81 | 1 | 95.2 | 97.4 | |
| TNLBTVenue=MMM’232026.04 | 81 | 1 | 95.2 | 97.4 | |
| Wang et al.Venue=MM’25, Citation Context=[51]2026.04 | 79.1 | 1 | 94.6 | 97 | |
| Wang et al.Venue=MM’252026.04 | 79.1 | 1 | 94.6 | 97 | |
| EOMAVenue=ICMR’222026.04 | 77.5 | 1 | 94.1 | 96.8 | |
| FMIVenue=Scientific Reports’252026.04 | 77.4 | 1 | 95.8 | 97.6 | |
| FMIVenue=Scientific Reports’252026.04 | 77.4 | 1 | 95.8 | 97.6 | |
| DARVenue=ECCV’242026.04 | 77.3 | 1 | 95.3 | 97.7 | |
| DARVenue=ECCV’242026.04 | 77.3 | 1 | 95.3 | 97.7 | |
| CIPVenue=MM’232026.04 | 77.1 | 1 | 94.2 | 97.2 | |
| CIPVenue=MM’232026.04 | 77.1 | 1 | 94.2 | 97.2 | |
| VLPCook (R1M+)Training Dataset=Recipe1M+2022.12 | 74.9 | 1 | 91.4 | 93.7 | |
| MALMVenue=arXiv’232026.04 | 74 | 1 | 91.3 | 94.3 | |
| FARMVenue=WACV’242026.04 | 73.7 | 1 | 90.7 | 93.4 | |
| FARMVenue=WACV’242026.04 | 73.7 | 1 | 90.7 | 93.4 | |
| VLPCook2022.12 | 73.6 | 1 | 90.5 | 93.3 | |
| VLPCookVenue=CVIU’242026.04 | 73.6 | 1 | 90.5 | 93.3 | |
| VLPCookVenue=CVIU’242026.04 | 73.6 | 1 | 90.5 | 93.3 | |
| CREAMYVenue=IEEE Access’242026.04 | 73.3 | 1 | 92.5 | 95.6 | |
| CREAMYVenue=IEEE Access’242026.04 | 73.3 | 1 | 92.5 | 95.6 | |
| T-Foodbackbone=CLIP-ViT2022.04 | 72.3 | 1 | 90.7 | 93.4 | |
| T-Food (CLIP-ViT)Backbone=CLIP-ViT2022.12 | 72.3 | 1 | 90.7 | 93.4 | |
| T-FoodVenue=CVPRW’222026.04 | 72.3 | 1 | 90.7 | 93.4 | |
| T-FoodVenue=CVPRW’222026.04 | 72.3 | 1 | 90.7 | 93.4 | |
| DCA-FoodVenue=Mathematics’242026.04 | 70.9 | 1 | 89.6 | 93.5 | |
| MMACMRVenue=Foods’242026.04 | 69.1 | 1 | 90.8 | 94.9 | |
| T-Foodbackbone=ViT2022.04 | 68.2 | 1 | 87.9 | 91.3 | |
| T-Food (ViT)Backbone=ViT2022.12 | 68.2 | 1 | 87.9 | 91.3 | |
| UTE-FCLVenue=Knowledge-Based Systems’242026.04 | 68.1 | 1 | 90 | 94.5 | |
| UTE-FCLVenue=Knowledge-Based Systems’242026.04 | 68.1 | 1 | 90 | 94.5 | |
| Papadopoulos et al.2022.12 | 66.9 | 1 | 90.9 | 95.1 | |
| Pap et al.Venue=CVPR’222026.04 | 66.9 | 1 | 90.9 | 95.1 | |
| LMF-CSFVenue=Multimedia Tools Appl.’242026.04 | 65.8 | 1 | 89.7 | 94.3 | |
| H-Tbackbone=ViT2022.04 | 64.2 | 1 | 89.1 | 93.4 | |
| H-T (ViT)Backbone=ViT2022.12 | 64.2 | 1 | 89.1 | 93.4 | |
| X-MRS2022.04 | 64 | 1 | 88.3 | 92.6 | |
| X-MRS2022.12 | 64 | 1 | 88.3 | 92.6 | |
| X-MRSVenue=MM’212026.04 | 64 | 1 | 88.3 | 92.6 | |
| X-MRSVenue=MM’212026.04 | 64 | 1 | 88.3 | 92.6 | |
| MFRFVenue=MMAsia’232026.04 | 62.3 | 1 | 86.9 | 91.5 | |
| Wang et al.Venue=MM’22, Citation Context=[47]2026.04 | 60.6 | — | 87.7 | 92.8 | |
| DaC2022.04 | 60.2 | 1 | 84 | 89.7 | |
| DaC2022.12 | 60.2 | 1 | 84 | 89.7 | |
| DaCVenue=arXiv’192026.04 | 60.2 | 1 | 84 | 89.7 | |
| H-T2022.04 | 60 | 1 | 87.6 | 92.9 | |
| H-TVenue=CVPR’212026.04 | 60 | 1 | 87.6 | 92.9 | |
| H-TVenue=CVPR’212026.04 | 60 | 1 | 87.6 | 92.9 | |
| RDE-GANVenue=MM’212026.04 | 59.4 | 1 | 81 | 87.4 | |
| LCWF-GIVenue=Connection Science’232026.04 | 59.4 | 1 | 86.8 | 92.5 | |
| RDE-GANVenue=MM’212026.04 | 59.4 | 1 | 81 | 87.4 | |
| M-SIA2022.04 | 59.3 | 1 | 86.3 | 92.6 | |
| M-SIA2022.12 | 59.3 | 1 | 86.3 | 92.6 | |
| M-SIAVenue=CIKM’212026.04 | 59.3 | 1 | 86.3 | 92.6 | |
| SEJE2022.04 | 58.1 | 1 | 85.8 | 92.2 | |
| SEJE2022.12 | 58.1 | 1 | 85.8 | 92.2 | |
| JEMAVenue=CIKM’212026.04 | 58.1 | 1 | 85.8 | 92.2 | |
| SEJEVenue=ACM Trans. Inf. Syst.’212026.04 | 58.1 | 1 | 85.8 | 92.2 | |
| MSJE2022.04 | 56.5 | 1 | 84.7 | 90.9 | |
| MSJE2022.12 | 56.5 | 1 | 84.7 | 90.9 | |
| MSJEVenue=IEEE TSC’222026.04 | 56.5 | 1 | 84.7 | 90.9 | |
| HF-ICMA2022.04 | 55.1 | 1 | 86.7 | 92.4 | |
| HF-ICMA2022.12 | 55.1 | 1 | 86.7 | 92.4 | |
| HF-ICMAVenue=SIGIR’212026.04 | 55.1 | 1 | 86.7 | 92.4 | |
| SCAN2022.04 | 54 | 1 | 81.7 | 88.8 | |
| SCAN2022.12 | 54 | 1 | 81.7 | 88.8 | |
| SCAN2020.03 | 54 | 1 | 81.9 | 89.2 | |
| SCANVenue=IEEE TMM’222026.04 | 54 | 1 | 81.9 | 89.2 | |
| Wang et al.2022.04 | 53.5 | 1 | 81.5 | 88.8 | |
| Wang et al.2022.12 | 53.5 | 1 | 81.5 | 88.8 | |
| Wang et al.Venue=IEEE TPAMI’23, Citation Context=[49]2026.04 | 53.5 | 1 | 81.5 | 88.8 | |
| IMHF2022.04 | 53.2 | 1 | 80.7 | 87.6 | |
| IMHF2022.12 | 53.2 | 1 | 80.7 | 87.6 | |
| IMHFVenue=ICMR’212026.04 | 53.2 | 1 | 80.7 | 87.6 | |
| SN2022.04 | 52.7 | 1 | 81.7 | 88.9 | |
| SN2022.12 | 52.7 | 1 | 81.7 | 88.9 | |
| tri-pro2020.03 | 52.7 | 1 | 81 | 88.1 | |
| SNVenue=ICMR’202026.04 | 52.7 | 1 | 81.7 | 88.9 | |
| ACME2022.04 | 51.8 | 1 | 80.2 | 87.5 | |
| ACME2022.12 | 51.8 | 1 | 80.2 | 87.5 | |
| ACME2020.03 | 51.8 | 1 | 80.2 | 87.5 | |
| ACMEVenue=CVPR’192026.04 | 51.8 | 1 | 80.2 | 87.5 | |
| ACMEVenue=CVPR’192026.04 | 51.8 | 1 | 80.2 | 87.5 | |
| CHEFVenue=AAAI’212026.04 | 49.4 | 1.8 | 79.6 | 86.1 | |
| MCEN2022.04 | 48.2 | 2 | 75.8 | 83.6 | |
| MCEN2022.12 | 48.2 | 2 | 75.8 | 83.6 | |
| MCEN2020.03 | 48.2 | 2 | 75.8 | 83.6 | |
| MCENVenue=CVPR’202026.04 | 48.2 | 2 | 75.8 | 83.6 | |
| MCENVenue=CVPR’202026.04 | 48.2 | 2 | 75.8 | 83.6 | |
| VLPCookPre-training status=without pretraining2022.12 | 45.2 | 2 | 75.9 | 84 | |
| Adamine2022.04 | 40.2 | 2 | 68.1 | 78.7 | |
| Adamine2022.12 | 40.2 | 2 | 68.1 | 78.7 | |
| AdaMine2020.03 | 39.8 | 2 | 69 | 77.4 |