Tasks SotA Datasets Papers Methods Submit About

Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable Benchmarks All SotA Datasets Papers Methods

Community

Submit Results About

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/ALBEF

ALBEF

Reported on 39 benchmarks across 11 tasks · 2 papers · 27 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Miscellaneous26 results

Image Retrieval with Multi-Modal QueryonCOCO 2014
Image-to-text R@1· uses extra data· 2021-07-16
77.6
best: 84.8 (BEiT-3)
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Image Retrieval with Multi-Modal QueryonCOCO 2014
Image-to-text R@10· uses extra data· 2021-07-16
97.2
best: 98.5 (X2-VLM (large))
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Image Retrieval with Multi-Modal QueryonCOCO 2014
Image-to-text R@5· uses extra data· 2021-07-16
94.3
best: 96.5 (X2-VLM (large))
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Image Retrieval with Multi-Modal QueryonCOCO 2014
Text-to-image R@1· uses extra data· 2021-07-16
60.7
best: 68 (VAST)
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Image Retrieval with Multi-Modal QueryonCOCO 2014
Text-to-image R@10· uses extra data· 2021-07-16
90.5
best: 92.8 (VAST)
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Image Retrieval with Multi-Modal QueryonCOCO 2014
Text-to-image R@5· uses extra data· 2021-07-16
84.3
best: 92.8 (BEiT-3)
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Cross-Modal Information RetrievalonCOCO 2014
Image-to-text R@1· uses extra data· 2021-07-16
77.6
best: 84.8 (BEiT-3)
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Cross-Modal Information RetrievalonCOCO 2014
Image-to-text R@10· uses extra data· 2021-07-16
97.2
best: 98.5 (X2-VLM (large))
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Cross-Modal Information RetrievalonCOCO 2014
Image-to-text R@5· uses extra data· 2021-07-16
94.3
best: 96.5 (X2-VLM (large))
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Cross-Modal Information RetrievalonCOCO 2014
Text-to-image R@1· uses extra data· 2021-07-16
60.7
best: 68 (VAST)
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Cross-Modal Information RetrievalonCOCO 2014
Text-to-image R@10· uses extra data· 2021-07-16
90.5
best: 92.8 (VAST)
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Cross-Modal Information RetrievalonCOCO 2014
Text-to-image R@5· uses extra data· 2021-07-16
84.3
best: 92.8 (BEiT-3)
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Image Retrieval with Multi-Modal QueryonCommercialAdsDataset
ADD(S) AUC· 2021-07-16
82.74
best: 91.73 (AlignCMSS)
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Image Retrieval with Multi-Modal QueryonFlickr30k
Image-to-text R@1· 2021-07-16
90.5
best: 98.8 (X2-VLM (large))
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Image Retrieval with Multi-Modal QueryonFlickr30k
Image-to-text R@10· 2021-07-16
99.7
best: 100 (X2-VLM (large))
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Image Retrieval with Multi-Modal QueryonFlickr30k
Image-to-text R@5· 2021-07-16
98.8
best: 100 (X2-VLM (large))
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Image Retrieval with Multi-Modal QueryonFlickr30k
Text-to-image R@1· 2021-07-16
76.8
best: 93.3 (ERNIE-ViL 2.0)
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Image Retrieval with Multi-Modal QueryonFlickr30k
Text-to-image R@10· 2021-07-16
96.7
best: 99.8 (ERNIE-ViL 2.0)
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Image Retrieval with Multi-Modal QueryonFlickr30k
Text-to-image R@5· 2021-07-16
93.7
best: 99.5 (M2-Encoder)
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Image Retrieval with Multi-Modal QueryonCOCO 2014
Image-to-text R@1· 2021-07-16
68.7
best: 84.8 (BEiT-3)
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Image Retrieval with Multi-Modal QueryonCOCO 2014
Image-to-text R@10· 2021-07-16
94.7
best: 98.5 (X2-VLM (large))
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Image Retrieval with Multi-Modal QueryonCOCO 2014
Image-to-text R@5· 2021-07-16
89.5
best: 96.5 (X2-VLM (large))
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Image Retrieval with Multi-Modal QueryonCOCO 2014
Text-to-image R@1· 2021-07-16
50.1
best: 68 (VAST)
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Image Retrieval with Multi-Modal QueryonCOCO 2014
Text-to-image R@10· 2021-07-16
84.5
best: 92.8 (VAST)
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Image Retrieval with Multi-Modal QueryonCOCO 2014
Text-to-image R@5· 2021-07-16
76.4
best: 92.8 (BEiT-3)
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Cross-Modal Information RetrievalonCommercialAdsDataset
ADD(S) AUC· 2021-07-16
82.74
best: 91.73 (AlignCMSS)
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651

Natural Language Processing13 results

Cross-Modal RetrievalonCOCO 2014
Image-to-text R@1· uses extra data· 2021-07-16
77.6
best: 84.8 (BEiT-3)
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Cross-Modal RetrievalonCOCO 2014
Image-to-text R@10· uses extra data· 2021-07-16
97.2
best: 98.5 (X2-VLM (large))
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Cross-Modal RetrievalonCOCO 2014
Image-to-text R@5· uses extra data· 2021-07-16
94.3
best: 96.5 (X2-VLM (large))
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Cross-Modal RetrievalonCOCO 2014
Text-to-image R@1· uses extra data· 2021-07-16
60.7
best: 68 (VAST)
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Cross-Modal RetrievalonCOCO 2014
Text-to-image R@10· uses extra data· 2021-07-16
90.5
best: 92.8 (VAST)
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Cross-Modal RetrievalonCOCO 2014
Text-to-image R@5· uses extra data· 2021-07-16
84.3
best: 92.8 (BEiT-3)
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Image-to-Text RetrievalonFlickr30k
Recall@1· 2021-07-16
95.9
best: 97.9 (InternVL-G-FT (finetuned, w/o ranking))
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Image-to-Text RetrievalonFlickr30k
Recall@10· 2021-07-16
100
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Image-to-Text RetrievalonFlickr30k
Recall@5· 2021-07-16
99.8
best: 100 (InternVL-G-FT (finetuned, w/o ranking))
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Image-to-Text RetrievalonFlickr30k
Recall@1· 2023-01-11
92.6
best: 97.9 (InternVL-G-FT (finetuned, w/o ranking))
HADA: A Graph-based Amalgamation Framework in Image-text Retrieval arXiv:2301.04742
Image-to-Text RetrievalonFlickr30k
Recall@10· 2023-01-11
99.9
best: 100 (InternVL-G-FT (finetuned, w/o ranking))
HADA: A Graph-based Amalgamation Framework in Image-text Retrieval arXiv:2301.04742
Image-to-Text RetrievalonFlickr30k
Recall@5· 2023-01-11
99.3
best: 100 (InternVL-G-FT (finetuned, w/o ranking))
HADA: A Graph-based Amalgamation Framework in Image-text Retrieval arXiv:2301.04742
Cross-Modal RetrievalonCommercialAdsDataset
ADD(S) AUC· 2021-07-16
82.74
best: 91.73 (AlignCMSS)
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651

Computer Vision5 results

Object DetectiononOVAD-Box benchmark
mean average precision· uses extra data· 2021-07-16
21
best: 28 (X-VLM)
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Open Vocabulary Object DetectiononOVAD-Box benchmark
mean average precision· uses extra data· 2021-07-16
21
best: 28 (X-VLM)
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Image RetrievalonFlickr30k
Recall@1· 2023-01-11
79.76
best: 89.7 (BLIP-2 ViT-G (zero-shot, 1K test set))
HADA: A Graph-based Amalgamation Framework in Image-text Retrieval arXiv:2301.04742
Image RetrievalonFlickr30k
Recall@10· 2023-01-11
97.72
best: 98.9 (BLIP-2 ViT-G (zero-shot, 1K test set))
HADA: A Graph-based Amalgamation Framework in Image-text Retrieval arXiv:2301.04742
Image RetrievalonFlickr30k
Recall@5· 2023-01-11
95.3
best: 98.1 (BLIP-2 ViT-G (zero-shot, 1K test set))
HADA: A Graph-based Amalgamation Framework in Image-text Retrieval arXiv:2301.04742

Methodology4 results

3DonOVAD-Box benchmark
mean average precision· uses extra data· 2021-07-16
21
best: 28 (X-VLM)
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
2D ClassificationonOVAD-Box benchmark
mean average precision· uses extra data· 2021-07-16
21
best: 28 (X-VLM)
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
2D Object DetectiononOVAD-Box benchmark
mean average precision· uses extra data· 2021-07-16
21
best: 28 (X-VLM)
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
16konOVAD-Box benchmark
mean average precision· uses extra data· 2021-07-16
21
best: 28 (X-VLM)
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651