HotpotQA
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
HotpotQA
93LLM Throughput (token/s)
8
HotpotQA Shuffle-Hotpot-Full
51.7F1 Score
8
HotpotQA Shuffle-Hotpot-Dist
55.9F1 Score
8
HotpotQA Full
64.2F1 Score
8
HotpotQA Hotpot-Dist
67.8F1 Score
8
HotPotQA
36Exact Match (HotPotQA)
8
HotpotQA
78.8Precision@5 (%)
8
HotPotQA Modified
27Average Correctness
8
HotPotQA Hard
0.13Top-1 ECE
8
HotpotQA (test)
100PoisonedRAG Success Rate
8
HotpotQA (test)
73.3AUROC
8
HotpotQA Covered Questions
0.839Accuracy
8
HotpotQA (All Questions)
80.2Accuracy
8
HotpotQA Normal
29.71F1 Score
8
HotpotQA (HOT) (test)
92.8Mean Answer Rate
8
HotpotQA
0.56Reward
8
HotpotQA
41.5F1 Score
8
HotpotQA
85.04nDCG@10
8
HotpotQA (val)
90.2R@2
8
HotpotQA
88.5Performance Score
8
HotpotQA
0.45ECE
8
HotpotQA
2,737Retrieved Tokens
8
HotpotQA
14.46Accuracy
8
HotpotQA
4.82Helpfulness
8
HotpotQA 1.0 (dev)
77.2EM
8