Few-NERD

TextsCC BY-SA 4.0Introduced 2021-05-16

Few-NERD is a large-scale, fine-grained manually annotated named entity recognition dataset, which contains 8 coarse-grained types, 66 fine-grained types, 188,200 sentences, 491,711 entities, and 4,601,223 tokens. Three benchmark tasks are built, one is supervised (Few-NERD (SUP)) and the other two are few-shot (Few-NERD (INTRA) and Few-NERD (INTER)).

Related Benchmarks

Few-NERD (INTER)/Named Entity Recognition (NER)/10 way 1~2 shot Few-NERD (INTER)/Named Entity Recognition (NER)/10 way 5~10 shot Few-NERD (INTER)/Named Entity Recognition (NER)/5 way 1~2 shot Few-NERD (INTER)/Named Entity Recognition (NER)/5 way 5~10 shot Few-NERD (INTER)/Named Entity Recognition (NER)/Average Few-NERD (INTRA)/Named Entity Recognition (NER)/10 way 1~2 shot Few-NERD (INTRA)/Named Entity Recognition (NER)/10 way 5~10 shot Few-NERD (INTRA)/Named Entity Recognition (NER)/5 way 1~2 shot Few-NERD (INTRA)/Named Entity Recognition (NER)/5 way 5~10 shot Few-NERD (INTRA)/Named Entity Recognition (NER)/Average Few-NERD (SUP)/Named Entity Recognition (NER)/F1-Measure Few-NERD (SUP)/Named Entity Recognition (NER)/Precision Few-NERD (SUP)/Named Entity Recognition (NER)/Recall