AMR Parsing on AMR 1.0 (test)
81.7SmatchStructBART-J
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| StructBART-JPre-trained Model=BART, Fine-tuning=true, Vocabulary Strategy=Joint2021.10 | 81.7 | — | — | — | |
| StructBART-SPre-trained Model=BART, Fine-tuning=true, Vocabulary Strategy=Separate2021.10 | 81.6 | — | — | — | |
| APT basebackbone=RoBERTa, decoding=partial ensemble2021.04 | 79.8 | — | — | — | |
| APT smallbackbone=RoBERTa, decoding=partial ensemble2021.04 | 79.7 | — | — | — | |
| APT basebackbone=RoBERTa2021.04 | 78.5 | — | — | — | |
| Zhou et al. (2021)Pre-trained Model=RoBERTa, Collapse Subgraph=S.A., External Dependency (NER)=true2021.10 | 78.3 | — | — | — | |
| mining + synTxt + synAMRsynthetic data source=unlabeled text2020.10 | 78.2 | — | — | — | |
| Lee et al.backbone=RoBERTa, training_data=85K silver2021.04 | 78.2 | — | — | — | |
| APT smallbackbone=RoBERTa2021.04 | 78.2 | — | — | — | |
| Lee et al. (2020)Pre-trained Model=RoBERTa, Collapse Subgraph=S.A., External Dependency (NER)=true, Extra Data=85K, Train Align.=true2021.10 | 78.2 | — | — | — | |
| synTxt + synAMRsynthetic data source=unlabeled text2020.10 | 78.1 | — | — | — | |
| mining + synAMRsynthetic data source=unlabeled text2020.10 | 77.7 | — | — | — | |
| synAMRsynthetic data source=unlabeled text2020.10 | 77.6 | — | — | — | |
| mining + synTxt2020.10 | 77.5 | — | — | — | |
| synTxt2020.10 | 77.3 | — | — | — | |
| F. A. et al. + SE^RBackbone=stack-Transformer2020.10 | 76.9 | — | — | — | |
| oracle mining2020.10 | 76.9 | — | — | — | |
| Astudillo et al.backbone=RoBERTa2021.04 | 76.9 | — | — | — | |
| Astudillo et al. (2020)Pre-trained Model=RoBERTa, Collapse Subgraph=S.A., External Dependency (NER)=true2021.10 | 76.9 | — | — | — | |
| Cai and LamGraph Recategorization=true, Embeddings=RoBERTa2020.10 | 75.4 | — | — | — | |
| Stack-TransformerAttention Modification=Full, Variant=c2020.10 | 75.4 | — | — | — | |
| F. A. et al.2020.10 | 75.4 | — | — | — | |
| Cai and LamBERT=base, Graph Recategorization=true2020.10 | 75.4 | — | — | — | |
| Cai and Lambackbone=BERT, graph re-categorization=true2021.04 | 75.4 | — | — | — | |
| Cai and Lam (2020)Pre-trained Model=BERT, Collapse Subgraph=G.R.2021.10 | 75.4 | — | — | — | |
| Stack-TransformerAttention Modification=buffer only, Variant=e2020.10 | 75.1 | — | — | — | |
| Cai and LamEmbeddings=BERT2020.10 | 74 | — | — | — | |
| TransformerMulti-task learning=true, Variant=b2020.10 | 74 | — | — | — | |
| Cai and LamBERT=base2020.10 | 74 | — | — | — | |
| Cai and Lambackbone=BERT2021.04 | 74 | — | — | — | |
| Cai and Lam (2020)Pre-trained Model=BERT2021.10 | 74 | — | — | — | |
| Lyu and TitovGraph Recategorization=true2020.10 | 73.7 | — | — | — | |
| Lyu and TitovGraph Recategorization=true2020.10 | 73.7 | — | — | — | |
| Zhang et al.Graph Recategorization=true, Embeddings=BERT2020.10 | 71.3 | — | — | — | |
| Zhang et al.BERT=large, Graph Recategorization=true2020.10 | 71.3 | — | — | — | |
| Zhang et al.backbone=BERT, graph re-categorization=true2021.04 | 71.3 | — | — | — | |
| Zhang et al. (2019b)Pre-trained Model=BERT, Collapse Subgraph=G.R., External Dependency (POS)=true, External Dependency (NER)=true, External Dependency (Lemm.)=true2021.10 | 71.3 | — | — | — | |
| Zhang et al.backbone=BERT, graph re-categorization=true2021.04 | 70.2 | — | — | — | |
| Zhang et al. (2019a)Pre-trained Model=BERT, Collapse Subgraph=G.R., External Dependency (POS)=true, External Dependency (NER)=true, External Dependency (Lemm.)=true2021.10 | 70.2 | — | — | — | |
| TransformerVariant=a2020.10 | 69.2 | — | — | — | |
| TransformerOracle=Naseem et al. (2019), Variant=a*2020.10 | 68.8 | — | — | — | |
| Guo and Lugraph re-categorization=true2021.04 | 68.3 | — | — | — | |
| Wang and Xue2021.04 | 68.1 | — | — | — | |
| Pust et al.2021.04 | 67.1 | — | — | — | |
| Flanigan et al.2021.04 | 66 | — | — | — | |
| BaselinePre-trained Model=None, Fine-Tune=None2020.10 | — | 69.8 | 60.2 | 64.6 | |
| Graph (Cai and Lam, 2020)Extra Resources=None2020.10 | — | — | — | 71.2 | |
| Graph (Cai and Lam, 2020)Extra Resources=BERT2020.10 | — | — | — | 75.4 | |
| PTM-MTPre-trained Model=PTM-MT, Fine-Tune=Vanilla2020.10 | — | 78.8 | 69.5 | 73.8 | |
| PTM-MTPre-trained Model=PTM-MT, Fine-Tune=MTL2020.10 | — | 81.1 | 72.2 | 76.4 | |
| PTM-MT-SemParPre-trained Model=PTM-MT-SemPar, Fine-Tune=Vanilla2020.10 | — | 82.3 | 75.4 | 78.7 | |
| PTM-MT-SemParPre-trained Model=PTM-MT-SemPar, Fine-Tune=MTL2020.10 | — | 82.4 | 74.6 | 78.3 | |
| PTM-MT-SynParPre-trained Model=PTM-MT-SynPar, Fine-Tune=Vanilla2020.10 | — | 79.1 | 70.5 | 74.6 | |
| PTM-MT-SynParPre-trained Model=PTM-MT-SynPar, Fine-Tune=MTL2020.10 | — | 81.2 | 74 | 77.5 | |
| PTM-MT-SynPar-SemParPre-trained Model=PTM-MT-SynPar-SemPar, Fine-Tune=Vanilla2020.10 | — | 82.4 | 75.4 | 78.7 | |
| PTM-MT-SynPar-SemParPre-trained Model=PTM-MT-SynPar-SemPar, Fine-Tune=MTL2020.10 | — | 82.6 | 75.9 | 79.1 | |
| PTM-SemParPre-trained Model=PTM-SemPar, Fine-Tune=Vanilla2020.10 | — | 80.8 | 73.5 | 77 | |
| PTM-SynParPre-trained Model=PTM-SynPar, Fine-Tune=Vanilla2020.10 | — | 74.3 | 65.8 | 69.8 | |
| PTM-SynParPre-trained Model=PTM-SynPar, Fine-Tune=MTL2020.10 | — | 76.7 | 68.1 | 72.2 | |
| PTM-SynPar-SemParPre-trained Model=PTM-SynPar-SemPar, Fine-Tune=Vanilla2020.10 | — | 81.6 | 74 | 77.6 | |
| PTM-SynPar-SemParPre-trained Model=PTM-SynPar-SemPar, Fine-Tune=MTL2020.10 | — | 81.8 | 74 | 77.7 | |
| Seq2Graph (Zhang et al., 2019a)Extra Resources=BERT2020.10 | — | — | — | 70.2 | |
| Seq2Graph (Zhang et al., 2019b)Extra Resources=BERT2020.10 | — | — | — | 71.3 |