ERNIE-ViL-large(ensemble of 15 models)

Reported on 3 benchmarks across 1 task · 1 paper · 3 SOTA

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

Natural Language Processing3 results

Visual Question Answering (VQA)onVCR (Q-AR) test
Accuracy· 2020-06-30
70.5
best: 81.6 (GPT4RoI)
SOTA
ERNIE-ViL: Knowledge Enhanced Vision-Language Representations Through Scene Graph arXiv:2006.16934
Visual Question Answering (VQA)onVCR (QA-R) test
Accuracy· 2020-06-30
86.1
best: 91 (GPT4RoI)
SOTA
ERNIE-ViL: Knowledge Enhanced Vision-Language Representations Through Scene Graph arXiv:2006.16934
Visual Question Answering (VQA)onVCR (Q-A) test
Accuracy· 2020-06-30
81.6
best: 89.4 (GPT4RoI)
SOTA
ERNIE-ViL: Knowledge Enhanced Vision-Language Representations Through Scene Graph arXiv:2006.16934