Visual Question Answering (VQA) 벤치마크
Visual Question Answering (VQA) on VCR (Q-A) dev
Accuracy
- 2019-08-09 — VisualBERT: Accuracy 70.8
- 2019-08-22 — VL-BERTLARGE: Accuracy 75.5
| Rank | Model | Accuracy | Paper | Code | Year |
|---|---|---|---|---|---|
| 1 | VL-BERTLARGE | 75.5 | VL-BERT: Pre-training of Generic Visual-Linguistic Representations | jackroos/VL-BERT · ImperialNLP/BertGen · jules-samaran/vl-bert | 2019 |
| 2 | VL-BERTBASE | 73.8 | VL-BERT: Pre-training of Generic Visual-Linguistic Representations | jackroos/VL-BERT · ImperialNLP/BertGen · jules-samaran/vl-bert | 2019 |
| 3 | VisualBERT | 70.8 | VisualBERT: A Simple and Performant Baseline for Vision and Language | uclanlp/visualbert · YIKUAN8/Transformers-VQA · lalithjets/surgical_vqa · +7 | 2019 |