HotpotQA
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
HotpotQA
5.46Inference Time (seconds)
4
HotpotQA
61.4EM
4
HotpotQA (500 randomly chosen questions)
1BM25 Queries
4
HotpotQA
425QPS
4
HotPotQA
89.9Precision
4
HotpotQA
95.6HS
4
HotpotQA (train test)
53.19BLEU
4
HotpotQA (sampled)
54Accuracy
4
HotpotQA
79.32Flans Score
4
HotpotQA (test)
0.453Recall@1
4
HotpotQA Fullwiki (test)
85.6Retrieval EM
4
KILT HotpotQA (test)
55.6Retrieval
4
HotpotQA (dev)
77.9Top-20 Accuracy
4
HotpotQA
39.6Recall
4
HotpotQA (test)
68.7EM
4
HotpotQA LongBench (test)
39String Match
3
HotpotQA bridge-k=2
63.2F1 Score
3
HotpotQA
68.1GPT-Acc
3
HotpotQA 400q
0Harm Rate
3
HotpotQA
92.5CDR
3
HotpotQA (test)
64.29Cost Reduction
3
HotpotQA (val)
23.56F1 Score
3
HotpotQA
34.68EM
3
HotpotQA Large-scale corpus (KILT)
53.4Exact Match (EM)
3
HotpotQA
90Score (Avg@3)
3