ProntoQA
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
ProntoQA (test)
99.72Accuracy
57
ProntoQA
99.2Pass@1
38
PrOntoQA 5-hop
90.2Accuracy
20
PRONTOQA (1,000 examples)
96.4Mean Hamming Similarity
20
PrOntoQA
91.4Calibrated Accuracy
17
ProntoQA
79.9AUROC
15
PrOntoQA
0.98Initial Score
15
PRONTOQA 5-hop
97.8Accuracy
14
ProntoQA
95Acc
14
ProntoQA (test)
2.8Error Rate
12
PrOntoQA
97.88PrOntoQA Score
10
PrOntoQA
97.3Accuracy (1 Hop)
8
PRONTOQA 4-hop
85Accuracy
6
PRONTOQA 3-hop
87Accuracy
6
ProntoQA (val)
98.01Accuracy
4
PRONTOQA 5-hop (test)
0.955Hamming Similarity
4
PRONTOQA 4-hop (test)
96.7Hamming Similarity
4
PRONTOQA 3-hop (test)
95.6Hamming Similarity
4
PrOntoQA
100Accuracy
3
ProntoQA Enhanced
99.8OA
1
ProntoQA OOD 500 records (test)
—ExcRate
0