Nested Named Entity Recognition on ACE 2005 (test)
90F1 ScoreBINDER
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| BINDEREncoder=BERT-large/BioBERT-large2022.08 | 90 | 89.6 | 90.5 | — | |
| BINDEREncoder=BERT-base/BioBERT2022.08 | 89.5 | 89.1 | 89.8 | — | |
| TriaffineEncoder=ALBERT2021.10 | 88.83 | 87.39 | 90.31 | — | |
| PIQNBackbone=bert-large2022.03 | 87.42 | 86.27 | 88.6 | — | |
| CNN_Nested_NERBackbone=ROBERTa-base, Parameters=125.6M, Data Source=This Work (New Preprocessing)2022.08 | 87.42 | 86.7 | 88.16 | — | |
| Shen et al. (2022)2022.10 | 87.42 | 86.27 | 88.6 | — | |
| DSpERTreporting_metric=best2022.10 | 87.42 | 87.01 | 87.84 | — | |
| CNN_Nested_NERBackbone=ROBERTa-base, Parameters=125.6M, Data Source=Zhu and Li (2022)2022.08 | 87.25 | 86.58 | 87.94 | — | |
| BSBackbone=ROBERTa-base, Parameters=125.6M, Data Source=This Work (New Preprocessing), reproduction=true2022.08 | 87.2 | 86.58 | 87.84 | — | |
| BSBackbone=ROBERTa-base, Parameters=125.6M, Data Source=Zhu and Li (2022)2022.08 | 87.15 | 86.25 | 88.07 | — | |
| Zhu and Li (2022)2022.10 | 87.15 | 86.25 | 88.07 | — | |
| Zhu and Li (2022)Category=Span-based, Backbone=bert-large2023.05 | 87.15 | 86.25 | 88.07 | — | |
| Zhu and Li (2022)Backbone=BERT2024.06 | 87.15 | — | — | — | |
| Tan et al.Encoder=BERT-large/BioBERT2022.08 | 87.1 | 87.5 | 86.6 | — | |
| Tan et al.Encoder=BERT2021.10 | 87.05 | 87.48 | 86.64 | — | |
| Seq2setcontext_sentences=true, pos_tags=true, data_preprocessing=different2022.03 | 87.05 | 87.48 | 86.63 | — | |
| Tan et al. (2021)2022.03 | 87.05 | 87.48 | 86.63 | — | |
| Sequence to SetEncoder=BERT2021.10 | 87.05 | 87.48 | 86.63 | — | |
| Tan et al. (2021)Category=Generation-based, Backbone=bert-large2023.05 | 87.05 | 87.48 | 86.63 | 84.91 | |
| Tan et al. (2021)Backbone=BERT2024.06 | 87.05 | — | — | — | |
| DSpERTreporting_metric=average2022.10 | 86.93 | 85.73 | 88.19 | — | |
| DIFFUSIONNERCategory=Diffusion-based, Backbone=bert-large2023.05 | 86.93 | 86.15 | 87.72 | 85.62 | |
| Nested NER as Latent Lexicalized Constituency Parsing2022.03 | 86.91 | 85.97 | 87.87 | — | |
| CNN_Nested_NERBackbone=BERT-large, Parameters=345.1M, Data Source=This Work (New Preprocessing)2022.08 | 86.91 | 86.26 | 87.56 | — | |
| Lou et al. (2022)Category=Span-based, Backbone=bert-large2023.05 | 86.91 | 85.97 | 87.87 | — | |
| Lou et al. (2022)Backbone=BERT2024.06 | 86.91 | — | — | — | |
| Li et al. (2020b)2022.03 | 86.88 | 87.16 | 86.59 | — | |
| Li et al. (2020b)2022.10 | 86.88 | 87.16 | 86.59 | — | |
| Li et al. (2020)Category=Span-based, Backbone=bert-large2023.05 | 86.88 | 87.16 | 86.59 | 83.92 | |
| TriaffineEncoder=BERT2021.10 | 86.82 | 86.7 | 86.94 | — | |
| CNN_Nested_NERBackbone=BERT-large, Parameters=345.1M, Data Source=Li et al. (2022)2022.08 | 86.82 | 86.39 | 87.24 | — | |
| Yuan et al. (2022b)Category=Span-based, Backbone=bert-large2023.05 | 86.82 | 86.7 | 86.94 | 85.14 | |
| W2NERBackbone=BERT-large, Parameters=355.4M, Data Source=Li et al. (2022)2022.08 | 86.79 | 85.03 | 88.62 | — | |
| Li et al. (2022)reporting_metric=average2022.10 | 86.79 | 85.03 | 88.62 | — | |
| Li et al. (2022a)Category=Span-based, Backbone=bert-large2023.05 | 86.79 | 85.03 | 88.62 | 85.23 | |
| W2NERBackbone=BERT-large, Parameters=355.4M, Data Source=This Work (New Preprocessing), reproduction=true2022.08 | 86.77 | 85.78 | 87.81 | — | |
| Locate and LabelBackbone=BERT + GloVe2021.05 | 86.67 | 86.09 | 87.27 | — | |
| Shen et al.Encoder=BERT2021.10 | 86.67 | 86.09 | 87.27 | — | |
| Locate&Labelcontext_sentences=true, pos_tags=true, data_preprocessing=different2022.03 | 86.67 | 86.09 | 87.27 | — | |
| Shen et al. (2021a)2022.03 | 86.67 | 86.09 | 87.27 | — | |
| Locate and LabelEncoder=BERT2021.10 | 86.67 | 86.09 | 87.27 | — | |
| Shen et al. (2021)2022.10 | 86.67 | 86.09 | 87.27 | — | |
| Shen et al. (2021)Category=Span-based, Backbone=bert-large2023.05 | 86.67 | 86.09 | 87.27 | 84.87 | |
| Shen et al. (2021)Backbone=BERT2024.06 | 86.67 | — | — | — | |
| CNN_Nested_NER (w.o. CNN)Backbone=ROBERTa-base, Parameters=125.2M, Data Source=This Work (New Preprocessing)2022.08 | 86.56 | 85.17 | 88.03 | — | |
| CNN_Nested_NER (w.o. CNN)Backbone=ROBERTa-base, Parameters=125.2M, Data Source=Zhu and Li (2022)2022.08 | 86.5 | 85.48 | 87.54 | — | |
| Pyramid-Fullextra_features=+ALBERT2022.03 | 86.34 | 85.3 | 87.4 | — | |
| PyramidEncoder=ALBERT2021.10 | 86.34 | 85.3 | 87.4 | — | |
| CNN_Nested_NER (w.o. CNN)Backbone=BERT-large, Parameters=343.6M, Data Source=This Work (New Preprocessing)2022.08 | 86.13 | 84.91 | 87.39 | — | |
| CNN_Nested_NER (w.o. CNN)Backbone=BERT-large, Parameters=343.6M, Data Source=Li et al. (2022)2022.08 | 85.92 | 84.88 | 86.99 | — | |
| Ours (8 bits)Backbone=BERT, Bits=82024.06 | 85.9 | — | — | — | |
| Locate&Labelrerun_on_current_data=true2022.03 | 85.82 | 86.02 | 85.62 | — | |
| Lu et al. (2022)Category=Generation-based, Backbone=bert-large2023.05 | 85.78 | — | — | — | |
| Ours (10 bits)Backbone=BERT, Bits=102024.06 | 85.75 | — | — | — | |
| Bottom-up Pointer NetworksEncoder=BERT2021.10 | 85.53 | 84.61 | 86.43 | — | |
| Yang and Tu (2022)2022.03 | 85.53 | 84.61 | 86.43 | — | |
| Yang and Tu (2022)Backbone=BERT2024.06 | 85.53 | — | — | — | |
| BoningKnifePre-trained model=BERT-base2021.07 | 85.46 | 84.77 | 86.16 | — | |
| Biaffine NER2020.05 | 85.4 | 85.2 | 85.6 | — | |
| Yu et al.2021.05 | 85.4 | 85.2 | 85.6 | — | |
| Fu et al.Encoder=BERT2021.10 | 85.4 | 84.5 | 86.4 | — | |
| Xu et al.Encoder=BERT2021.10 | 85.4 | 85.7 | 85.2 | — | |
| PO-TreeCRFsdata_preprocessing=different2022.03 | 85.4 | 84.5 | 86.4 | — | |
| Yu et al. (2020)2022.03 | 85.4 | 85.2 | 85.6 | — | |
| TreeCRFEncoder=BERT2021.10 | 85.4 | 84.5 | 86.4 | — | |
| BiaffineEncoder=BERT2021.10 | 85.4 | 85.2 | 85.6 | — | |
| Yu et al.Encoder=BERT-large/BioBERT2022.08 | 85.4 | 85.2 | 85.6 | — | |
| Yu et al. (2020)2022.10 | 85.4 | 85.2 | 85.6 | — | |
| Yu et al. (2020)Category=Span-based, Backbone=bert-large2023.05 | 85.4 | 85.2 | 85.6 | 84.2 | |
| Xu et al. (2021)Backbone=BERT2024.06 | 85.4 | — | — | — | |
| Fu et al. (2021)Backbone=BERT2024.06 | 85.4 | — | — | — | |
| Yu et al. (2020)Backbone=BERT2024.06 | 85.4 | — | — | — | |
| Ours (12 bits)Backbone=BERT, Bits=122024.06 | 85.26 | — | — | — | |
| Ours (6 bits)Backbone=BERT, Bits=62024.06 | 85.23 | — | — | — | |
| CNN_Nested_NERBackbone=BERT-base, Parameters=110.5M, Data Source=Wan et al. (2022)2022.08 | 85.16 | 84.94 | 85.4 | — | |
| SGBackbone=BERT-base, Parameters=112.3M, Data Source=Wan et al. (2022)2022.08 | 85.11 | 84.37 | 85.87 | — | |
| Wan et al. (2022)Category=Span-based, Backbone=bert-large2023.05 | 85.11 | 84.37 | 85.87 | 83.57 | |
| Wan et al.Encoder=BERT-base2022.08 | 85.1 | 84.4 | 85.9 | — | |
| Bottom-up Pointer Networks (w.o. E_yt-1)Encoder=BERT, Eq.1=w.o. E_yt-12021.10 | 85.01 | 83.75 | 86.31 | — | |
| Zhang et al.Encoder=T5-base2022.08 | 84.9 | 83.3 | 86.6 | — | |
| Yan et al.Encoder=BERT2021.10 | 84.74 | 83.16 | 86.38 | — | |
| BARTNER2022.03 | 84.74 | 83.16 | 86.38 | — | |
| Yan et al. (2021)2022.03 | 84.74 | 83.16 | 86.38 | — | |
| BartNEREncoder=BART2021.10 | 84.74 | 83.16 | 86.38 | — | |
| Yan et al. (2021)2022.10 | 84.74 | 83.16 | 86.38 | — | |
| Yan et al. (2021b)Category=Generation-based, Backbone=bert-large2023.05 | 84.74 | 83.16 | 86.38 | 83.6 | |
| Wang et al. (2021)Encoder=BERT2021.10 | 84.71 | 85.28 | 84.15 | — | |
| W(max)2022.03 | 84.71 | 85.28 | 84.15 | — | |
| Wang et al. (2021)Backbone=BERT2024.06 | 84.71 | — | — | — | |
| Wang et al.Encoder=BERT-large/BioBERT2022.08 | 84.7 | 84 | 85.4 | — | |
| Yan et al.Encoder=BART-large2022.08 | 84.7 | 83.2 | 86.4 | — | |
| Wang et al.suffix=2020a2021.05 | 84.66 | 83.95 | 85.39 | — | |
| Wang et al. (2020)Encoder=BERT2021.10 | 84.66 | 83.95 | 85.39 | — | |
| Pyramid-Basic2022.03 | 84.66 | 83.95 | 85.39 | — | |
| Wang et al. (2020a)2022.03 | 84.66 | 83.95 | 85.39 | — | |
| PyramidEncoder=BERT2021.10 | 84.66 | 83.95 | 85.39 | — | |
| Wang et al. (2020)Category=Tagging-based, Backbone=bert-large2023.05 | 84.66 | 83.95 | 85.39 | 83.57 | |
| Wang et al. (2020)Backbone=BERT2024.06 | 84.66 | — | — | — | |
| Li et al.Encoder=BERT-large/BioBERT2022.08 | 84.6 | 85 | 84.1 | — | |
| PO-TreeCRFsrerun_on_current_data=true2022.03 | 84.49 | 83.34 | 85.67 | — |