2WikiMQA
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
2WikiMQA
76.4F1 Score
175
2WikiMQA
74.9F1
66
2WikiMQA
79.5SubEM
36
2WikiMQA (test)
44.3Unreliability Rate
28
2WikiMQA IRCoT 500 samples (test)
52.8ACC
27
2WikiMQA
55.47F1-Recall
22
2WikiMQA (Passage Split)
52.53Score
18
2WikiMQA Fixed Chunk 2048
52.53QA Score
18
2WikiMQA LongEmbed
96.5nDCG@10
9
2WikiMQA (test)
69.7Recall@K
8
2WikiMQA (test)
48.6Exact Match
7
2WikiMQA
80.2NDCG@5
5
2WikiMQA
7Joint Answer-Evidence Score
4
2WikiMQA (sampled)
0.63Accuracy
4