HotpotQA
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
HotpotQA
43.1EM
3
HotpotQA Search-o1 framework (test)
28.8Accuracy
3
HotpotQA (val)
0.804AUROC
3
HotpotQA Long (2.1k tok)
43.3Exact Match (EM)
3
HotPotQA (test)
53.5P@1
3
HotpotQA transductive (val)
91.8R@5
3
HotpotQA distractor (val)
78P@1
3
HotPotQA
82.7Accuracy
3
HotPotQA
48.06Token F1
3
HotpotQA
45.4Accuracy
3
HotpotQA
90.9Accuracy (Context Extraction)
3
HotpotQA MultiTurn
27ASR
3
HotpotQA 300 Questions
31.3Exact Match (EM)
3
HotpotQA multi-hop (Noisy)
31.5Precision
3
HotpotQA (Clean)
32.9Precision
3
HotpotQA 1,000 samples
0.626EM
3
HotpotQA
85.24Accuracy
3
HotpotQA
72.88Answer EM
3
HotpotQA
72.88Answer EM
3
HotpotQA 16k sequence length (test)
72.88Answer EM
3
HotpotQA 12k sequence length (test)
73.5Answer EM
3
HotpotQA 8k sequence length (test)
74.1Answer EM
3
HotpotQA
69.5EM
3
HotpotQA
86.3ROC-AUC
3
HotPotQA 10-shot
67.08Accuracy
3