Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

EntityQuestions

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multi-source information fusion in RAG-based Web Q&AEntityQuestions
AUC Score538.31
72
Question AnsweringEntityQuestions
Accuracy69.5
20
Knowledge RecoveryEntityQuestions (test)
Steered Accuracy36
18
Question AnsweringEntityQuestions Frequent
Exact Match Accuracy38.7
16
Question AnsweringEntityQuestions (Infrequent)
Exact Match (EM)38.6
16
Question AnsweringEntityQuestions (Long-tail)
Exact Match Accuracy36.4
16
Open-Domain Question Answering RetrievalEntityQuestions (test)
Accuracy@2076.3
15
RetrievalENTITYQUESTIONS Infrequent
Recall@1075.4
14
Information RetrievalEntityQuestions (test)
Recall@20 (macro average)74.5
11
Passage RetrievalEntityQuestions unseen query distribution (test)
Accuracy@200.662
8
RetrievalENTITYQUESTIONS Frequent
Recall@1071.5
7
RetrievalENTITYQUESTIONS Long-tail
Recall@1054.6
7
RetrievalENTITYQUESTIONS (Frequent)
Recall@1081.9
7
RetrievalENTITYQUESTIONS Long-tail
Recall@100.726
7
Open-domain Question AnsweringEntityQuestions (ENTITYQ) (test)
EM28.3
7
Information RetrievalEntityQuestions
Acc@2079.8
4
Showing 16 of 16 rows