ALBEF (14M)

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.

Natural Language Processing2 results

Visual Question Answering (VQA)onVQA v2 test-dev
Accuracy· 2021-07-16
75.84
best: 84.3 (PaLI)
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Visual Question Answering (VQA)onVQA v2 test-std
overall· 2021-07-16
76.04
best: 84.03 (BEiT-3)
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651

Reasoning2 results

Visual ReasoningonNLVR2 Dev
Accuracy· 2021-07-16
83.14
best: 91.51 (BEiT-3)
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651
Visual ReasoningonNLVR2 Test
Accuracy· 2021-07-16
82.55
best: 92.58 (BEiT-3)
SOTA
Align before Fuse: Vision and Language Representation Learning with Momentum Distillation arXiv:2107.07651