Recursive Dense Retriever

Reported on 6 benchmarks across 1 task · 1 paper · 2 SOTA

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
JOINT-EM· 2020-09-27
0.418
best: 0.505 (Beam Retrieval)
SOTA
Answering Complex Open-Domain Questions with Multi-Hop Dense Retrieval arXiv:2009.12756
Question AnsweringonHotpotQA
SUP-EM· 2020-09-27
0.575
best: 0.663 (Beam Retrieval)
SOTA
Answering Complex Open-Domain Questions with Multi-Hop Dense Retrieval arXiv:2009.12756
Question AnsweringonHotpotQA
ANS-EM· 2020-09-27
0.623
best: 0.727 (Beam Retrieval)
Answering Complex Open-Domain Questions with Multi-Hop Dense Retrieval arXiv:2009.12756
Question AnsweringonHotpotQA
ANS-F1· 2020-09-27
0.753
best: 0.85 (Beam Retrieval)
Answering Complex Open-Domain Questions with Multi-Hop Dense Retrieval arXiv:2009.12756
Question AnsweringonHotpotQA
JOINT-F1· 2020-09-27
0.666
best: 0.775 (Beam Retrieval)
Answering Complex Open-Domain Questions with Multi-Hop Dense Retrieval arXiv:2009.12756
Question AnsweringonHotpotQA
SUP-F1· 2020-09-27
0.809
best: 0.901 (Beam Retrieval)
Answering Complex Open-Domain Questions with Multi-Hop Dense Retrieval arXiv:2009.12756