Co-Tokenization

Reported on 3 benchmarks across 2 tasks · 1 paper

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)onMSRVTT-QA
Accuracy· uses extra data· 2022-08-01
0.457
best: 0.496 (VLAB)
Video Question Answering with Iterative Video-Text Co-Tokenization arXiv:2208.00934
Visual Question Answering (VQA)onMSVD-QA
Accuracy· uses extra data· 2022-08-01
0.486
best: 0.61 (VLAB)
Video Question Answering with Iterative Video-Text Co-Tokenization arXiv:2208.00934

Reasoning1 result

Video Question AnsweringoniVQA
Accuracy· 2022-08-01
38.2
best: 40.2 (Text + Text (no Multimodal Pretext Training))
Video Question Answering with Iterative Video-Text Co-Tokenization arXiv:2208.00934