2WikiMHQA
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
2WikiMHQA
85.56F1 Score
85
2WikiMHQA
0.7AUROC
26
2WikiMHQA Wiki18 index (test)
55.1Pass@4
20
2WikiMHQA (test)
71.85EM
17
2WikiMHQA in-distribution
79.39Exact Match (EM)
17
2WikiMHQA (test)
45.8EM
14
2WikiMHQA
42.1Exact Match (EM)
12
2WikiMHQA
10.8Query Latency (s/q)
8
2WikiMHQA 500 samples (test)
91.1F1 Score
8
2WikiMHQA Covered Questions
66.9Accuracy
8
2WikiMHQA All Questions
61Accuracy
8
2WikiMHQA
37.8EM
7
2WikiMHQA
0.56Average Violation
4
2WikiMHQA
72.3GPT-Acc
3
2WikiMHQA
71.2Accuracy
2
2WikiMHQA KET-RAG
64.8Accuracy
2
2WikiMHQA strict (test)
0.592Base Error
1
2WikiMHQA adaptive (test)
0.311Base Error
1
2WikiMHQA
0.311Base Error
1
2WikiMHQA in-distribution v4 (test)
—EM
0