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/X2-VLM (base)

X2-VLM (base)

Reported on 51 benchmarks across 8 tasks · 1 paper

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· 2022-11-22
98.5
best: 98.8 (X2-VLM (large))
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Image Retrieval with Multi-Modal QueryonFlickr30k
Image-to-text R@10· uses extra data· 2022-11-22
100
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Image Retrieval with Multi-Modal QueryonFlickr30k
Image-to-text R@5· uses extra data· 2022-11-22
100
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Image Retrieval with Multi-Modal QueryonFlickr30k
Text-to-image R@1· uses extra data· 2022-11-22
90.4
best: 93.3 (ERNIE-ViL 2.0)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Image Retrieval with Multi-Modal QueryonFlickr30k
Text-to-image R@10· uses extra data· 2022-11-22
99.3
best: 99.8 (ERNIE-ViL 2.0)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Image Retrieval with Multi-Modal QueryonFlickr30k
Text-to-image R@5· uses extra data· 2022-11-22
98.2
best: 99.5 (M2-Encoder)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Image Retrieval with Multi-Modal QueryonCOCO 2014
Image-to-text R@1· uses extra data· 2022-11-22
83.5
best: 84.8 (BEiT-3)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Image Retrieval with Multi-Modal QueryonCOCO 2014
Image-to-text R@10· uses extra data· 2022-11-22
98.5
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Image Retrieval with Multi-Modal QueryonCOCO 2014
Image-to-text R@5· uses extra data· 2022-11-22
96.3
best: 96.5 (X2-VLM (large))
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Image Retrieval with Multi-Modal QueryonCOCO 2014
Text-to-image R@1· uses extra data· 2022-11-22
66.2
best: 68 (VAST)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Image Retrieval with Multi-Modal QueryonCOCO 2014
Text-to-image R@10· uses extra data· 2022-11-22
92.2
best: 92.8 (VAST)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Image Retrieval with Multi-Modal QueryonCOCO 2014
Text-to-image R@5· uses extra data· 2022-11-22
87.1
best: 92.8 (BEiT-3)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Cross-Modal Information RetrievalonFlickr30k
Image-to-text R@1· uses extra data· 2022-11-22
98.5
best: 98.8 (X2-VLM (large))
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Cross-Modal Information RetrievalonFlickr30k
Image-to-text R@10· uses extra data· 2022-11-22
100
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Cross-Modal Information RetrievalonFlickr30k
Image-to-text R@5· uses extra data· 2022-11-22
100
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Cross-Modal Information RetrievalonFlickr30k
Text-to-image R@1· uses extra data· 2022-11-22
90.4
best: 93.3 (ERNIE-ViL 2.0)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Cross-Modal Information RetrievalonFlickr30k
Text-to-image R@10· uses extra data· 2022-11-22
99.3
best: 99.8 (ERNIE-ViL 2.0)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Cross-Modal Information RetrievalonFlickr30k
Text-to-image R@5· uses extra data· 2022-11-22
98.2
best: 99.4 (ERNIE-ViL 2.0)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Cross-Modal Information RetrievalonCOCO 2014
Image-to-text R@1· uses extra data· 2022-11-22
83.5
best: 84.8 (BEiT-3)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Cross-Modal Information RetrievalonCOCO 2014
Image-to-text R@10· uses extra data· 2022-11-22
98.5
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Cross-Modal Information RetrievalonCOCO 2014
Image-to-text R@5· uses extra data· 2022-11-22
96.3
best: 96.5 (X2-VLM (large))
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Cross-Modal Information RetrievalonCOCO 2014
Text-to-image R@1· uses extra data· 2022-11-22
66.2
best: 68 (VAST)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Cross-Modal Information RetrievalonCOCO 2014
Text-to-image R@10· uses extra data· 2022-11-22
92.2
best: 92.8 (VAST)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Cross-Modal Information RetrievalonCOCO 2014
Text-to-image R@5· uses extra data· 2022-11-22
87.1
best: 92.8 (BEiT-3)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402

Natural Language Processing16 results

Visual Question Answering (VQA)onMSRVTT-QA
Accuracy· uses extra data· 2022-11-22
0.45
best: 0.496 (VLAB)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Visual Question Answering (VQA)onMSVD-QA
Accuracy· uses extra data· 2022-11-22
0.528
best: 0.61 (VLAB)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Visual Question Answering (VQA)onVQA v2 test-dev
Accuracy· 2022-11-22
80.4
best: 84.3 (PaLI)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Visual Question Answering (VQA)onVQA v2 test-std
overall· 2022-11-22
80.2
best: 84.03 (BEiT-3)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Cross-Modal RetrievalonFlickr30k
Image-to-text R@1· uses extra data· 2022-11-22
98.5
best: 98.8 (X2-VLM (large))
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Cross-Modal RetrievalonFlickr30k
Image-to-text R@10· uses extra data· 2022-11-22
100
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Cross-Modal RetrievalonFlickr30k
Image-to-text R@5· uses extra data· 2022-11-22
100
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Cross-Modal RetrievalonFlickr30k
Text-to-image R@1· uses extra data· 2022-11-22
90.4
best: 93.3 (ERNIE-ViL 2.0)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Cross-Modal RetrievalonFlickr30k
Text-to-image R@10· uses extra data· 2022-11-22
99.3
best: 99.8 (ERNIE-ViL 2.0)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Cross-Modal RetrievalonFlickr30k
Text-to-image R@5· uses extra data· 2022-11-22
98.2
best: 99.4 (ERNIE-ViL 2.0)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Cross-Modal RetrievalonCOCO 2014
Image-to-text R@1· uses extra data· 2022-11-22
83.5
best: 84.8 (BEiT-3)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Cross-Modal RetrievalonCOCO 2014
Image-to-text R@10· uses extra data· 2022-11-22
98.5
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Cross-Modal RetrievalonCOCO 2014
Image-to-text R@5· uses extra data· 2022-11-22
96.3
best: 96.5 (X2-VLM (large))
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Cross-Modal RetrievalonCOCO 2014
Text-to-image R@1· uses extra data· 2022-11-22
66.2
best: 68 (VAST)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Cross-Modal RetrievalonCOCO 2014
Text-to-image R@10· uses extra data· 2022-11-22
92.2
best: 92.8 (VAST)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Cross-Modal RetrievalonCOCO 2014
Text-to-image R@5· uses extra data· 2022-11-22
87.1
best: 92.8 (BEiT-3)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402

Computer Vision9 results

VideoonMSR-VTT-1kA
text-to-video R@1· 2022-11-22
47.6
best: 62.9 (HunYuan_tvr (huge))
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
VideoonMSR-VTT-1kA
text-to-video R@10· 2022-11-22
84.2
best: 90.8 (HunYuan_tvr (huge))
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
VideoonMSR-VTT-1kA
text-to-video R@5· 2022-11-22
74.1
best: 84.5 (HunYuan_tvr (huge))
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Video RetrievalonMSR-VTT-1kA
text-to-video R@1· 2022-11-22
47.6
best: 62.9 (HunYuan_tvr (huge))
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Video RetrievalonMSR-VTT-1kA
text-to-video R@10· 2022-11-22
84.2
best: 90.8 (HunYuan_tvr (huge))
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Video RetrievalonMSR-VTT-1kA
text-to-video R@5· 2022-11-22
74.1
best: 84.5 (HunYuan_tvr (huge))
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Visual GroundingonRefCOCO+ test B
Accuracy (%)· 2022-11-22
78.4
best: 92 (Florence-2-large-ft)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Visual GroundingonRefCOCO+ val
Accuracy (%)· 2022-11-22
85.2
best: 93.4 (Florence-2-large-ft)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Visual GroundingonRefCOCO+ testA
Accuracy (%)· 2022-11-22
90.3
best: 95.3 (Florence-2-large-ft)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402

Reasoning2 results

Visual ReasoningonNLVR2 Dev
Accuracy· 2022-11-22
86.2
best: 91.51 (BEiT-3)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402
Visual ReasoningonNLVR2 Test
Accuracy· 2022-11-22
87
best: 92.58 (BEiT-3)
X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks arXiv:2211.12402