Recipe-to-Image Retrieval on Recipe1M 1k setup (test)
85.1Recall@1SIMMER
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| SIMMERVersion=V1-7B2026.04 | 85.1 | 1 | 97.6 | 99.1 | |
| SIMMER (V1-7B)Venue=-2026.04 | 85.1 | 1 | 97.6 | 99.1 | |
| SIMMERVersion=V22026.04 | 81.7 | 1 | 96.5 | 98.3 | |
| SIMMER (V2)Venue=-2026.04 | 81.7 | 1 | 96.5 | 98.3 | |
| SIMMERVersion=V1-2B2026.04 | 81.5 | 1 | 96.6 | 98.4 | |
| SIMMER (V1-2B)Venue=-2026.04 | 81.5 | 1 | 96.6 | 98.4 | |
| Yang et al.Venue=ICMRW’242026.04 | 81.2 | 1 | 96 | 97.9 | |
| Yang et al.Venue=ICMRW’242026.04 | 81.2 | 1 | 96 | 97.9 | |
| TNLBTVenue=MMM’232026.04 | 80.3 | 1 | 95.2 | 97.4 | |
| TNLBTVenue=MMM’232026.04 | 80.3 | 1 | 95.2 | 97.4 | |
| Wang et al.Venue=MM’25, Citation Context=[51]2026.04 | 78.3 | 1 | 95 | 97.2 | |
| Wang et al.Venue=MM’252026.04 | 78.3 | 1 | 95 | 97.2 | |
| EOMAVenue=ICMR’222026.04 | 77.5 | 1 | 94 | 96.7 | |
| CIPVenue=MM’232026.04 | 77.3 | 1 | 94.4 | 97 | |
| CIPVenue=MM’232026.04 | 77.3 | 1 | 94.4 | 97 | |
| DARVenue=ECCV’242026.04 | 77.1 | 1 | 95.4 | 97.9 | |
| FMIVenue=Scientific Reports’252026.04 | 77.1 | 1 | 95.4 | 97.7 | |
| DARVenue=ECCV’242026.04 | 77.1 | 1 | 95.4 | 97.9 | |
| FMIVenue=Scientific Reports’252026.04 | 77.1 | 1 | 95.4 | 97.7 | |
| VLPCook (R1M+)Training Dataset=Recipe1M+2022.12 | 75.6 | 1 | 91.2 | 93.6 | |
| VLPCook2022.12 | 74.7 | 1 | 90.7 | 93.2 | |
| VLPCookVenue=CVIU’242026.04 | 74.7 | 1 | 90.7 | 93.2 | |
| VLPCookVenue=CVIU’242026.04 | 74.7 | 1 | 90.7 | 93.2 | |
| FARMVenue=WACV’242026.04 | 73.6 | 1 | 90.8 | 93.5 | |
| FARMVenue=WACV’242026.04 | 73.6 | 1 | 90.8 | 93.5 | |
| CREAMYVenue=IEEE Access’242026.04 | 73.2 | 1 | 92.5 | 95.8 | |
| CREAMYVenue=IEEE Access’242026.04 | 73.2 | 1 | 92.5 | 95.8 | |
| MALMVenue=arXiv’232026.04 | 73 | 1 | 91 | 93.9 | |
| T-Foodbackbone=CLIP-ViT2022.04 | 72.6 | 1 | 90.6 | 93.4 | |
| T-Food (CLIP-ViT)Backbone=CLIP-ViT2022.12 | 72.6 | 1 | 90.6 | 93.4 | |
| T-FoodVenue=CVPRW’222026.04 | 72.6 | 1 | 90.6 | 93.4 | |
| T-FoodVenue=CVPRW’222026.04 | 72.6 | 1 | 90.6 | 93.4 | |
| DCA-FoodVenue=Mathematics’242026.04 | 70.8 | 1 | 90.1 | 93.9 | |
| MMACMRVenue=Foods’242026.04 | 69.2 | 1 | 90.6 | 95 | |
| T-Foodbackbone=ViT2022.04 | 68.3 | 1 | 87.8 | 91.5 | |
| T-Food (ViT)Backbone=ViT2022.12 | 68.3 | 1 | 87.8 | 91.5 | |
| UTE-FCLVenue=Knowledge-Based Systems’242026.04 | 67.2 | 1 | 90.1 | 94.7 | |
| UTE-FCLVenue=Knowledge-Based Systems’242026.04 | 67.2 | 1 | 90.1 | 94.7 | |
| Papadopoulos et al.2022.12 | 66.8 | 1 | 89.8 | 94.6 | |
| Pap et al.Venue=CVPR’222026.04 | 66.8 | 1 | 89.8 | 94.6 | |
| LMF-CSFVenue=Multimedia Tools Appl.’242026.04 | 65.5 | 1 | 89.4 | 94.3 | |
| MFRFVenue=MMAsia’232026.04 | 64.8 | 1 | 88.7 | 92.7 | |
| H-Tbackbone=ViT2022.04 | 64.5 | 1 | 89.3 | 93.8 | |
| H-T (ViT)Backbone=ViT2022.12 | 64.5 | 1 | 89.3 | 93.8 | |
| X-MRS2022.04 | 63.9 | 1 | 87.6 | 92.6 | |
| X-MRS2022.12 | 63.9 | 1 | 87.6 | 92.6 | |
| X-MRSVenue=MM’212026.04 | 63.9 | 1 | 87.6 | 92.6 | |
| X-MRSVenue=MM’212026.04 | 63.9 | 1 | 87.6 | 92.6 | |
| Wang et al.Venue=MM’22, Citation Context=[47]2026.04 | 61.3 | — | 87.7 | 93.2 | |
| RDE-GANVenue=MM’212026.04 | 61.2 | 1 | 81 | 87.2 | |
| RDE-GANVenue=MM’212026.04 | 61.2 | 1 | 81 | 87.2 | |
| H-T2022.04 | 60.3 | 1 | 87.6 | 93.2 | |
| H-TVenue=CVPR’212026.04 | 60.3 | 1 | 87.6 | 93.2 | |
| H-TVenue=CVPR’212026.04 | 60.3 | 1 | 87.6 | 93.2 | |
| LCWF-GIVenue=Connection Science’232026.04 | 60.1 | 1 | 86.7 | 92.7 | |
| M-SIA2022.04 | 59.8 | 1 | 86.7 | 92.8 | |
| M-SIA2022.12 | 59.8 | 1 | 86.7 | 92.8 | |
| M-SIAVenue=CIKM’212026.04 | 59.8 | 1 | 86.7 | 92.8 | |
| SEJE2022.04 | 58.5 | 1 | 86.2 | 92.3 | |
| SEJE2022.12 | 58.5 | 1 | 86.2 | 92.3 | |
| JEMAVenue=CIKM’212026.04 | 58.5 | 1 | 86.2 | 92.3 | |
| SEJEVenue=ACM Trans. Inf. Syst.’212026.04 | 58.5 | 1 | 86.2 | 92.3 | |
| HF-ICMA2022.04 | 56.8 | 1 | 87.5 | 93 | |
| HF-ICMA2022.12 | 56.8 | 1 | 87.5 | 93 | |
| HF-ICMAVenue=SIGIR’212026.04 | 56.8 | 1 | 87.5 | 93 | |
| MSJE2022.04 | 56.2 | 1 | 84.9 | 91.1 | |
| MSJE2022.12 | 56.2 | 1 | 84.9 | 91.1 | |
| MSJEVenue=IEEE TSC’222026.04 | 56.2 | 1 | 84.9 | — | |
| Wang et al.2022.04 | 55 | 1 | 82 | 88.8 | |
| Wang et al.2022.12 | 55 | 1 | 82 | 88.8 | |
| Wang et al.Venue=IEEE TPAMI’23, Citation Context=[49]2026.04 | 55 | 1 | 82 | 88.8 | |
| SCAN2022.04 | 54.9 | 1 | 81.9 | 89 | |
| SCAN2022.12 | 54.9 | 1 | 81.9 | 89 | |
| SCAN2020.03 | 54.9 | 1 | 81.9 | 89 | |
| SCANVenue=IEEE TMM’222026.04 | 54.9 | 1 | 81.9 | 89 | |
| SN2022.04 | 54.1 | 1 | 81.8 | 88.9 | |
| IMHF2022.04 | 54.1 | 1 | 82.4 | 88.2 | |
| SN2022.12 | 54.1 | 1 | 81.8 | 88.9 | |
| IMHF2022.12 | 54.1 | 1 | 82.4 | 88.2 | |
| SNVenue=ICMR’202026.04 | 54.1 | 1 | 81.8 | 88.9 | |
| IMHFVenue=ICMR’212026.04 | 54.1 | 1 | 82.4 | 88.2 | |
| tri-pro2020.03 | 53.8 | 1 | 81.1 | 88.3 | |
| ACME2022.04 | 52.8 | 1 | 80.2 | 87.6 | |
| ACME2022.12 | 52.8 | 1 | 80.2 | 87.6 | |
| ACME2020.03 | 52.8 | 1 | 80.2 | 87.6 | |
| ACMEVenue=CVPR’192026.04 | 52.8 | 1 | 80.2 | 87.6 | |
| ACMEVenue=CVPR’192026.04 | 52.8 | 1 | 80.2 | 87.6 | |
| CHEFVenue=AAAI’212026.04 | 49.8 | 1.5 | 79 | 86.4 | |
| MCEN2022.04 | 48.4 | 1.9 | 76.1 | 83.7 | |
| MCEN2022.12 | 48.4 | 1.9 | 76.1 | 83.7 | |
| MCEN2020.03 | 48.4 | 2 | 76.1 | 83.7 | |
| MCENVenue=CVPR’202026.04 | 48.4 | 1.9 | 76.1 | 83.7 | |
| MCENVenue=CVPR’202026.04 | 48.4 | 1.9 | 76.1 | 83.7 | |
| R2GAN2022.04 | 40.6 | 2 | 72.6 | 83.3 | |
| R2GAN2022.12 | 40.6 | 2 | 72.6 | 83.3 | |
| R2GAN2020.03 | 40.6 | 2 | 72.6 | 83.3 | |
| R2GANVenue=CVPR’192026.04 | 40.6 | 2 | 72.6 | 83.3 | |
| AdaMine2020.03 | 40.2 | 1 | 68.1 | 78.7 | |
| AdaMineVenue=SIGIR’182026.04 | 40.2 | 1 | 68.1 | 78.7 | |
| AdaMineVenue=SIGIR’182026.04 | 40.2 | 1 | 68.1 | 78.7 |