HotpotQA
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
HotpotQA
67.6Avg@4
9
HotpotQA Full Document Context (test)
23.33BLEU-4
9
HotpotQA v1.0 (test)
2.28Latency (s)
9
HotpotQA
33.91Accuracy
9
HotpotQA
61.8Citation Recall
9
HotpotQA Adversarial
56.8EM
9
HotpotQA official Wikipedia paragraphs
67.1EM
9
HotpotQA
72.9Evidence Index (IE)
8
HotpotQA
69.1Accuracy
8
HotpotQA original questions as queries BEIR
46.9Accuracy
8
HotpotQA
69.74Precision
8
HotpotQA
42.3Accuracy
8
HotpotQA
54.7Exact Match (EM)
8
HotpotQA
0FDR
8
HotpotQA
79.69Score (7k)
8
HotpotQA
47.1F1 Score
8
HotpotQA
1.18Avg Edge Multiplicity
8
HotpotQA
14Query Latency (s/q)
8
HotpotQA 500 samples (test)
83.9F1 Score
8
HotpotQA bridge
35.13F1 Score
8
HotPotQA
57.6Task Score
8
HotpotQA Full context
60.8Accuracy
8
HotpotQA Proxy context
93.2Accuracy
8
HotpotQA
86.69F1 Score
8
HotpotQA
0.032Retrieval Latency (s)
8