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/VALOR

VALOR

Reported on 73 benchmarks across 12 tasks · 1 paper · 39 SOTA

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

Computer Vision43 results

VideoonVATEX
text-to-video R@1· uses extra data· 2023-04-17
78.5
best: 87.7 (GRAM)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
VideoonVATEX
text-to-video R@10· uses extra data· 2023-04-17
98.7
best: 100 (GRAM)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
VideoonVATEX
text-to-video R@5· uses extra data· 2023-04-17
97.1
best: 98.2 (VAST)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
VideoonActivityNet
text-to-video R@1· uses extra data· 2023-04-17
70.1
best: 74.1 (InternVideo2-6B)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
VideoonActivityNet
text-to-video R@10· uses extra data· 2023-04-17
95.3
best: 96.1 (GRAM)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
VideoonActivityNet
text-to-video R@5· uses extra data· 2023-04-17
90.8
best: 90.9 (VAST)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
VideoonMSR-VTT
text-to-video R@1· uses extra data· 2023-04-17
59.9
best: 64 (GRAM)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
VideoonMSR-VTT
text-to-video R@10· uses extra data· 2023-04-17
89.6
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
VideoonMSR-VTT
text-to-video R@5· uses extra data· 2023-04-17
83.5
best: 84.3 (VAST)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video CaptioningonVATEX
BLEU-4· uses extra data· 2023-04-17
45.6
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video CaptioningonVATEX
CIDEr· uses extra data· 2023-04-17
95.8
best: 99.5 (VAST)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video CaptioningonVATEX
METEOR· uses extra data· 2023-04-17
29.4
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video CaptioningonVATEX
ROUGE-L· uses extra data· 2023-04-17
57.4
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video CaptioningonMSVD
BLEU-4· uses extra data· 2023-04-17
80.7
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video CaptioningonMSVD
METEOR· uses extra data· 2023-04-17
51
best: 51.2 (VLAB)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video CaptioningonMSVD
ROUGE-L· uses extra data· 2023-04-17
87.9
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video RetrievalonVATEX
text-to-video R@1· uses extra data· 2023-04-17
78.5
best: 87.7 (GRAM)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video RetrievalonVATEX
text-to-video R@10· uses extra data· 2023-04-17
98.7
best: 100 (GRAM)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video RetrievalonVATEX
text-to-video R@5· uses extra data· 2023-04-17
97.1
best: 98.2 (VAST)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video RetrievalonActivityNet
text-to-video R@1· uses extra data· 2023-04-17
70.1
best: 74.1 (InternVideo2-6B)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video RetrievalonActivityNet
text-to-video R@10· uses extra data· 2023-04-17
95.3
best: 96.1 (GRAM)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video RetrievalonActivityNet
text-to-video R@5· uses extra data· 2023-04-17
90.8
best: 90.9 (VAST)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video RetrievalonMSR-VTT
text-to-video R@1· uses extra data· 2023-04-17
59.9
best: 64 (GRAM)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video RetrievalonMSR-VTT
text-to-video R@10· uses extra data· 2023-04-17
89.6
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video RetrievalonMSR-VTT
text-to-video R@5· uses extra data· 2023-04-17
83.5
best: 84.3 (VAST)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Audio-visual Question AnsweringonMUSIC-AVQA
Acc· uses extra data· 2023-04-17
78.9
best: 80.7 (VAST)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
VideoonDiDeMo
text-to-video R@1· uses extra data· 2023-04-17
61.5
best: 74.2 (InternVideo2-6B)
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
VideoonDiDeMo
text-to-video R@10· uses extra data· 2023-04-17
90.4
best: 94.2 (vid-TLDR (UMT-L))
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
VideoonDiDeMo
text-to-video R@5· uses extra data· 2023-04-17
85.3
best: 91.2 (vid-TLDR (UMT-L))
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
VideoonLSMDC
text-to-video R@1· uses extra data· 2023-04-17
34.2
best: 46.4 (InternVideo2-6B)
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
VideoonLSMDC
text-to-video R@10· uses extra data· 2023-04-17
64.1
best: 92.8 (HunYuan_tvr (huge))
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
VideoonLSMDC
text-to-video R@5· uses extra data· 2023-04-17
56
best: 80.1 (HunYuan_tvr (huge))
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video CaptioningonMSR-VTT
BLEU-4· uses extra data· 2023-04-17
54.4
best: 57.8 (mPLUG-2)
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video CaptioningonMSR-VTT
CIDEr· uses extra data· 2023-04-17
74
best: 80 (mPLUG-2)
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video CaptioningonMSR-VTT
METEOR· uses extra data· 2023-04-17
32.9
best: 38.7 (MV-GPT)
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video CaptioningonMSR-VTT
ROUGE-L· uses extra data· 2023-04-17
68
best: 70.1 (mPLUG-2)
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video CaptioningonMSVD
CIDEr· uses extra data· 2023-04-17
178.5
best: 195.6 (MaMMUT)
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video RetrievalonDiDeMo
text-to-video R@1· uses extra data· 2023-04-17
61.5
best: 74.2 (InternVideo2-6B)
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video RetrievalonDiDeMo
text-to-video R@10· uses extra data· 2023-04-17
90.4
best: 94.2 (vid-TLDR (UMT-L))
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video RetrievalonDiDeMo
text-to-video R@5· uses extra data· 2023-04-17
85.3
best: 91.2 (vid-TLDR (UMT-L))
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video RetrievalonLSMDC
text-to-video R@1· uses extra data· 2023-04-17
34.2
best: 46.4 (InternVideo2-6B)
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video RetrievalonLSMDC
text-to-video R@10· uses extra data· 2023-04-17
64.1
best: 92.8 (HunYuan_tvr (huge))
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video RetrievalonLSMDC
text-to-video R@5· uses extra data· 2023-04-17
56
best: 80.1 (HunYuan_tvr (huge))
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345

Audio14 results

Audio captioningonClotho
BLEU-4· uses extra data· 2023-04-17
16.2
best: 19 (VAST)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Audio captioningonClotho
CIDEr· uses extra data· 2023-04-17
0.423
best: 14 (ZerAuCap)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Audio captioningonClotho
METEOR· uses extra data· 2023-04-17
17.4
best: 19.3 (VAST)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Audio captioningonClotho
ROUGE-L· uses extra data· 2023-04-17
38.2
best: 40.8 (VAST)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Audio captioningonAudioCaps
BLEU-4· uses extra data· 2023-04-17
0.27
best: 14.3 (Audio Flamingo)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Audio captioningonAudioCaps
METEOR· uses extra data· 2023-04-17
0.231
best: 20.5 (Audio Flamingo)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Audio captioningonAudioCaps
ROUGE-L· uses extra data· 2023-04-17
0.494
best: 40.8 (Audio Flamingo)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Text to Audio RetrievalonAudioCaps
R@1· uses extra data· 2023-04-17
40.1
best: 55.2 (InternVideo2-6B)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Text to Audio RetrievalonAudioCaps
R@5· uses extra data· 2023-04-17
73.9
best: 77.5 (ONE-PEACE)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Text to Audio RetrievalonClotho
R@1· uses extra data· 2023-04-17
17.5
best: 27.69 (PaSST-RoBERTa & Estimated Audio–Caption Correspondences)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Text to Audio RetrievalonClotho
R@10· uses extra data· 2023-04-17
55.3
best: 70.39 (PaSST-RoBERTa & Estimated Audio–Caption Correspondences)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Text to Audio RetrievalonClotho
R@5· uses extra data· 2023-04-17
42.7
best: 57.03 (PaSST-RoBERTa & Estimated Audio–Caption Correspondences)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Audio captioningonAudioCaps
CIDEr· uses extra data· 2023-04-17
0.741
best: 50.2 (Audio Flamingo)
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Text to Audio RetrievalonAudioCaps
R@10· uses extra data· 2023-04-17
83.1
best: 88.4 (ONE-PEACE)
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345

Natural Language Processing8 results

Visual Question Answering (VQA)onMSVD-QA
Accuracy· uses extra data· 2023-04-17
0.6
best: 0.61 (VLAB)
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Visual Question Answering (VQA)onVQA v2 test-dev
Accuracy· uses extra data· 2023-04-17
78.46
best: 84.3 (PaLI)
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Visual Question Answering (VQA)onVQA v2 test-std
overall· uses extra data· 2023-04-17
78.62
best: 84.03 (BEiT-3)
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Image CaptioningonCOCO Captions
CIDER· uses extra data· 2023-04-17
152.5
best: 155.1 (mPLUG)
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Image CaptioningonCOCO Captions
SPICE· uses extra data· 2023-04-17
25.7
best: 27 (VAST)
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Cross-Modal RetrievalonCOCO 2014
Text-to-image R@1· uses extra data· 2023-04-17
61.4
best: 68 (VAST)
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Cross-Modal RetrievalonCOCO 2014
Text-to-image R@10· uses extra data· 2023-04-17
90.9
best: 92.8 (VAST)
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Cross-Modal RetrievalonCOCO 2014
Text-to-image R@5· uses extra data· 2023-04-17
84.4
best: 92.8 (BEiT-3)
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345

Miscellaneous6 results

Image Retrieval with Multi-Modal QueryonCOCO 2014
Text-to-image R@1· uses extra data· 2023-04-17
61.4
best: 68 (VAST)
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Image Retrieval with Multi-Modal QueryonCOCO 2014
Text-to-image R@10· uses extra data· 2023-04-17
90.9
best: 92.8 (VAST)
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Image Retrieval with Multi-Modal QueryonCOCO 2014
Text-to-image R@5· uses extra data· 2023-04-17
84.4
best: 92.8 (BEiT-3)
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Cross-Modal Information RetrievalonCOCO 2014
Text-to-image R@1· uses extra data· 2023-04-17
61.4
best: 68 (VAST)
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Cross-Modal Information RetrievalonCOCO 2014
Text-to-image R@10· uses extra data· 2023-04-17
90.9
best: 92.8 (VAST)
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Cross-Modal Information RetrievalonCOCO 2014
Text-to-image R@5· uses extra data· 2023-04-17
84.4
best: 92.8 (BEiT-3)
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345

Reasoning2 results

Video Question AnsweringonMSRVTT-QA
Accuracy· uses extra data· 2023-04-17
49.2
best: 72.4 (Flash-VStream)
SOTA
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345
Video Question AnsweringonActivityNet-QA
Accuracy· uses extra data· 2023-04-17
48.6
best: 61.6 (Tarsier (34B))
VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset arXiv:2304.08345