WebQA
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
WebQA
99.02ASR
64
WebQA Average
50.8F1 Score
32
WebQA
73.57AUROC
30
WebQA
90.92F1-Recall
22
WebQA
43.55Recall@20
21
WebQA
76.52Recall@20
21
WebQA
84.7Recall@5
19
WebQA Text-grounded n=2,455 (test)
48.6F1 Score
16
WebQA Image-grounded n=2,511 (test)
69.4F1 Score
16
WebQA Text
54.9F1 Score
16
WebQA Image
13.8F1 Score
16
WebQA (IID)
100Recall
16
WebQA NIID-1
99.12Recall
16
WebQA
1.05Rank
16
WebQA
53.9Accuracy
14
WebQA (test)
0.623BLEURT
14
WebQA
59.3ROrig
12
WebQA
100Recall
12
WebQA image segment 1.0 (test)
49.8Accuracy
12
WebQA k=2
64.8ROrig@k
8
WebQA (200 held-out questions)
62.7Accuracy
7
WebQA+
40.96Recall@1
7
WebQA CN
57.15Accuracy
6
WebQA 3 (test)
79.5Accuracy (%)
6
WebQA 2
83.15R@5
6