Just Ask (fine-tune)

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.

Reasoning3 results

Video Question AnsweringonActivityNet-QA
Accuracy· 2020-12-01
38.9
best: 61.6 (Tarsier (34B))
SOTA
Just Ask: Learning to Answer Questions from Millions of Narrated Videos arXiv:2012.00451
Video Question AnsweringoniVQA
Accuracy· 2020-12-01
35.4
best: 40.2 (Text + Text (no Multimodal Pretext Training))
SOTA
Just Ask: Learning to Answer Questions from Millions of Narrated Videos arXiv:2012.00451
Video Question AnsweringonVideoQA
Accuracy· 2020-12-01
15.6
SOTA
Just Ask: Learning to Answer Questions from Millions of Narrated Videos arXiv:2012.00451