List your company

ResearchFormerly WizWand

2Wiki

Benchmarks

TaskDatasetSOTA resultTrendResults

Question Answering

2WIKI

86EM

260

Multi-hop Question Answering

2Wiki

74.9Exact Match

215

Question Answering

2Wiki (test)

61.8EM Accuracy

59

Class-level Continual Learning

2Wiki

71.74Average Accuracy (AA)

56

Multi-Hop Question Answering

2Wiki

43.6Accuracy (2Wiki)

44

2Wiki

62EM

42

2Wiki

96.13Recall@5

42

Multi-hop Question Answering

2Wiki (test)

69.7F1 Score

34

Multi-hop QA Retrieval

2Wiki

90.1Recall@2

32

Adversarial Attack

2Wiki

78.8ASR

30

Multi-Hop Question Answering

2Wiki (out-of-domain)

47.6Accuracy

28

Question Answering

2Wiki 100K context

78.91Accuracy

25

Multi-Hop Question Answering

2Wiki (test)

45.1Accuracy

24

Multi-Hop Search-augmented Question Answering

2Wiki

45.2Success Rate

22

Multi-hop Question Answering

2Wiki 18

43.6Exact Match (EM)

20

Multi-Hop Question Answering

2Wiki

58.7Token-Level F1

20

Question Answering

2Wiki 30K context

73.7Accuracy

19

Question Answering

2Wiki 10K context

72.2Accuracy

19

Multi-hop Question Answering

2Wiki

58pass@1

18

Multi-Hop Question Answering

2Wiki

46.9Exact Match (EM)

18

Multi-Hop Question Answering

2wiki

61.3F1 Score

17

2Wiki

70.5Accuracy

17

Multi-hop Question Answering

2Wiki

59MBE

17

Multi-hop Question Answering

2Wiki

48.89EM

16

Multi-Hop QA Verification

2wiki

81.21P@1

16