Quark + SemanticRetrievalMRS IR

Reported on 6 benchmarks across 1 task · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing6 results

Question AnsweringonHotpotQA
ANS-EM· 2020-04-14
0.555
best: 0.727 (Beam Retrieval)
A Simple Yet Strong Pipeline for HotpotQA arXiv:2004.06753
Question AnsweringonHotpotQA
ANS-F1· 2020-04-14
0.675
best: 0.85 (Beam Retrieval)
A Simple Yet Strong Pipeline for HotpotQA arXiv:2004.06753
Question AnsweringonHotpotQA
JOINT-EM· 2020-04-14
0.329
best: 0.505 (Beam Retrieval)
A Simple Yet Strong Pipeline for HotpotQA arXiv:2004.06753
Question AnsweringonHotpotQA
JOINT-F1· 2020-04-14
0.562
best: 0.775 (Beam Retrieval)
A Simple Yet Strong Pipeline for HotpotQA arXiv:2004.06753
Question AnsweringonHotpotQA
SUP-EM· 2020-04-14
0.456
best: 0.663 (Beam Retrieval)
A Simple Yet Strong Pipeline for HotpotQA arXiv:2004.06753
Question AnsweringonHotpotQA
SUP-F1· 2020-04-14
0.73
best: 0.901 (Beam Retrieval)
A Simple Yet Strong Pipeline for HotpotQA arXiv:2004.06753