SOHO

Reported on 4 benchmarks across 2 tasks · 1 paper · 3 SOTA

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

Reasoning2 results

Visual ReasoningonNLVR2 Dev
Accuracy· 2021-04-07
76.37
best: 91.51 (BEiT-3)
SOTA
Seeing Out of tHe bOx: End-to-End Pre-training for Vision-Language Representation Learning arXiv:2104.03135
Visual ReasoningonNLVR2 Test
Accuracy· 2021-04-07
77.32
best: 92.58 (BEiT-3)
Seeing Out of tHe bOx: End-to-End Pre-training for Vision-Language Representation Learning arXiv:2104.03135

Natural Language Processing2 results

Natural Language InferenceonSNLI-VE val
Accuracy· 2021-04-07
85
best: 91 (OFA)
SOTA
Seeing Out of tHe bOx: End-to-End Pre-training for Vision-Language Representation Learning arXiv:2104.03135
Natural Language InferenceonSNLI-VE test
Accuracy· 2021-04-07
84.95
best: 91.2 (OFA)
SOTA
Seeing Out of tHe bOx: End-to-End Pre-training for Vision-Language Representation Learning arXiv:2104.03135