OTT-QA
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
OTT-QA (test)
96.7Recall@10
27
OTT-QA 1.0 (dev)
56.9EM
27
OTT-QA (test)
54.9EM
14
OTT-QA in-distribution (test)
37.44Average@32
11
OTT-QA 1.0 (test)
43EM
10
OTT-QA (dev)
57.6Table Recall R@1
8
OTT-QA (dev)
40EM
8
OTT-QA
0.317Accuracy
6
OTT-QA full (val)
56.9EM
5
OTT-QA
47.3Precision
3
OTT-QA (dev)
80.5Top-20 Acc
3
OTT-QA sampled 500 questions (val)
48.8EM
1