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/X-VLM (base)

X-VLM (base)

Reported on 47 benchmarks across 8 tasks · 1 paper · 36 SOTA

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

Miscellaneous24 results

Image Retrieval with Multi-Modal QueryonFlickr30k
Image-to-text R@1· uses extra data· 2021-11-16
97.1
best: 98.8 (X2-VLM (large))
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Image Retrieval with Multi-Modal QueryonFlickr30k
Image-to-text R@5· uses extra data· 2021-11-16
100
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Image Retrieval with Multi-Modal QueryonFlickr30k
Text-to-image R@1· uses extra data· 2021-11-16
86.9
best: 93.3 (ERNIE-ViL 2.0)
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Image Retrieval with Multi-Modal QueryonFlickr30k
Text-to-image R@10· uses extra data· 2021-11-16
98.7
best: 99.8 (ERNIE-ViL 2.0)
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Image Retrieval with Multi-Modal QueryonCOCO 2014
Image-to-text R@1· uses extra data· 2021-11-16
81.2
best: 84.8 (BEiT-3)
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Image Retrieval with Multi-Modal QueryonCOCO 2014
Image-to-text R@10· uses extra data· 2021-11-16
98.2
best: 98.5 (X2-VLM (large))
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Image Retrieval with Multi-Modal QueryonCOCO 2014
Image-to-text R@5· uses extra data· 2021-11-16
95.6
best: 96.5 (X2-VLM (large))
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Image Retrieval with Multi-Modal QueryonCOCO 2014
Text-to-image R@1· uses extra data· 2021-11-16
63.4
best: 68 (VAST)
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Image Retrieval with Multi-Modal QueryonCOCO 2014
Text-to-image R@10· uses extra data· 2021-11-16
91.5
best: 92.8 (VAST)
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Image Retrieval with Multi-Modal QueryonCOCO 2014
Text-to-image R@5· uses extra data· 2021-11-16
85.8
best: 92.8 (BEiT-3)
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Cross-Modal Information RetrievalonFlickr30k
Image-to-text R@1· uses extra data· 2021-11-16
97.1
best: 98.8 (X2-VLM (large))
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Cross-Modal Information RetrievalonFlickr30k
Image-to-text R@5· uses extra data· 2021-11-16
100
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Cross-Modal Information RetrievalonFlickr30k
Text-to-image R@1· uses extra data· 2021-11-16
86.9
best: 93.3 (ERNIE-ViL 2.0)
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Cross-Modal Information RetrievalonFlickr30k
Text-to-image R@10· uses extra data· 2021-11-16
98.7
best: 99.8 (ERNIE-ViL 2.0)
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Cross-Modal Information RetrievalonCOCO 2014
Image-to-text R@1· uses extra data· 2021-11-16
81.2
best: 84.8 (BEiT-3)
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Cross-Modal Information RetrievalonCOCO 2014
Image-to-text R@10· uses extra data· 2021-11-16
98.2
best: 98.5 (X2-VLM (large))
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Cross-Modal Information RetrievalonCOCO 2014
Image-to-text R@5· uses extra data· 2021-11-16
95.6
best: 96.5 (X2-VLM (large))
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Cross-Modal Information RetrievalonCOCO 2014
Text-to-image R@1· uses extra data· 2021-11-16
63.4
best: 68 (VAST)
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Cross-Modal Information RetrievalonCOCO 2014
Text-to-image R@10· uses extra data· 2021-11-16
91.5
best: 92.8 (VAST)
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Cross-Modal Information RetrievalonCOCO 2014
Text-to-image R@5· uses extra data· 2021-11-16
85.8
best: 92.8 (BEiT-3)
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Image Retrieval with Multi-Modal QueryonFlickr30k
Image-to-text R@10· uses extra data· 2021-11-16
100
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Image Retrieval with Multi-Modal QueryonFlickr30k
Text-to-image R@5· uses extra data· 2021-11-16
97.3
best: 99.5 (M2-Encoder)
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Cross-Modal Information RetrievalonFlickr30k
Image-to-text R@10· uses extra data· 2021-11-16
100
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Cross-Modal Information RetrievalonFlickr30k
Text-to-image R@5· uses extra data· 2021-11-16
97.3
best: 99.4 (ERNIE-ViL 2.0)
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276

Natural Language Processing15 results

Cross-Modal RetrievalonFlickr30k
Image-to-text R@1· uses extra data· 2021-11-16
97.1
best: 98.8 (X2-VLM (large))
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Cross-Modal RetrievalonFlickr30k
Image-to-text R@5· uses extra data· 2021-11-16
100
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Cross-Modal RetrievalonFlickr30k
Text-to-image R@1· uses extra data· 2021-11-16
86.9
best: 93.3 (ERNIE-ViL 2.0)
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Cross-Modal RetrievalonFlickr30k
Text-to-image R@10· uses extra data· 2021-11-16
98.7
best: 99.8 (ERNIE-ViL 2.0)
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Cross-Modal RetrievalonCOCO 2014
Image-to-text R@1· uses extra data· 2021-11-16
81.2
best: 84.8 (BEiT-3)
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Cross-Modal RetrievalonCOCO 2014
Image-to-text R@10· uses extra data· 2021-11-16
98.2
best: 98.5 (X2-VLM (large))
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Cross-Modal RetrievalonCOCO 2014
Image-to-text R@5· uses extra data· 2021-11-16
95.6
best: 96.5 (X2-VLM (large))
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Cross-Modal RetrievalonCOCO 2014
Text-to-image R@1· uses extra data· 2021-11-16
63.4
best: 68 (VAST)
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Cross-Modal RetrievalonCOCO 2014
Text-to-image R@10· uses extra data· 2021-11-16
91.5
best: 92.8 (VAST)
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Cross-Modal RetrievalonCOCO 2014
Text-to-image R@5· uses extra data· 2021-11-16
85.8
best: 92.8 (BEiT-3)
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Visual Question Answering (VQA)onVQA v2 test-dev
Accuracy· 2021-11-16
78.22
best: 84.3 (PaLI)
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Image CaptioningonCOCO Captions
BLEU-4· 2021-11-16
41.3
best: 46.5 (mPLUG)
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Image CaptioningonCOCO Captions
CIDER· 2021-11-16
140.8
best: 155.1 (mPLUG)
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Cross-Modal RetrievalonFlickr30k
Image-to-text R@10· uses extra data· 2021-11-16
100
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Cross-Modal RetrievalonFlickr30k
Text-to-image R@5· uses extra data· 2021-11-16
97.3
best: 99.4 (ERNIE-ViL 2.0)
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276

Computer Vision6 results

Image RetrievalonFlickr30K 1K test
R@1· uses extra data· 2021-11-16
86.9
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Image RetrievalonFlickr30K 1K test
R@10· uses extra data· 2021-11-16
98.7
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Image RetrievalonFlickr30K 1K test
R@5· uses extra data· 2021-11-16
97.3
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Visual GroundingonRefCOCO+ test B
Accuracy (%)· 2021-11-16
76.91
best: 92 (Florence-2-large-ft)
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Visual GroundingonRefCOCO+ val
Accuracy (%)· 2021-11-16
84.51
best: 93.4 (Florence-2-large-ft)
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Visual GroundingonRefCOCO+ testA
Accuracy (%)· 2021-11-16
89
best: 95.3 (Florence-2-large-ft)
SOTA
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276

Reasoning2 results

Visual ReasoningonNLVR2 Dev
Accuracy· 2021-11-16
84.41
best: 91.51 (BEiT-3)
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276
Visual ReasoningonNLVR2 Test
Accuracy· 2021-11-16
84.76
best: 92.58 (BEiT-3)
Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts arXiv:2111.08276