Just Ask

Reported on 5 benchmarks across 3 tasks · 1 paper · 5 SOTA

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

Natural Language Processing4 results

Visual Question Answering (VQA)onMSVD-QA
Accuracy· 2020-12-01
0.463
best: 0.61 (VLAB)
SOTA
Just Ask: Learning to Answer Questions from Millions of Narrated Videos arXiv:2012.00451
Visual Question Answering (VQA)onMSRVTT-QA
Accuracy· 2020-12-01
0.415
best: 0.496 (VLAB)
SOTA
Just Ask: Learning to Answer Questions from Millions of Narrated Videos arXiv:2012.00451
Visual Question AnsweringonMSVD-QA
Accuracy· 2020-12-01
0.463
best: 0.548 (FrozenBiLM)
SOTA
Just Ask: Learning to Answer Questions from Millions of Narrated Videos arXiv:2012.00451
Visual Question AnsweringonMSRVTT-QA
Accuracy· 2020-12-01
0.415
best: 0.47 (FrozenBiLM)
SOTA
Just Ask: Learning to Answer Questions from Millions of Narrated Videos arXiv:2012.00451

Reasoning1 result

Video Question AnsweringonHow2QA
Accuracy· uses extra data· 2020-12-01
84.4
best: 93.2 (Text + Text (no Multimodal Pretext Training))
SOTA
Just Ask: Learning to Answer Questions from Millions of Narrated Videos arXiv:2012.00451