Argument Component Identification on CMV Modes (test)
0.5Claim PrecisionBase-LF
Evaluation results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Base-LFBackbone=Longformer2022.03 | 0.5 | 0.5 | 0.5 | 0.58 | 0.64 | 0.61 | 0.56 | 0.74 | |
| SMLM-LFBackbone=Longformer, Pre-training Strategy=Selective Masked Language Modeling2022.03 | 0.49 | 0.57 | 0.53 | 0.61 | 0.67 | 0.64 | 0.59 | 0.74 | |
| SMLM-ROBERTaBackbone=RoBERTa, Pre-training Strategy=Selective Masked Language Modeling2022.03 | 0.49 | 0.6 | 0.53 | 0.55 | 0.57 | 0.55 | 0.55 | 0.72 | |
| ROBERTaBackbone=RoBERTa2022.03 | 0.49 | 0.55 | 0.51 | 0.56 | 0.62 | 0.59 | 0.56 | 0.73 | |
| BERTBackbone=BERT2022.03 | 0.21 | 0.25 | 0.23 | 0.19 | 0.26 | 0.22 | 0.22 | 0.62 | |
| LSTM-MDataBackbone=BiLSTM-CNN-CRF2022.03 | 0.19 | 0.18 | 0.18 | 0.26 | 0.23 | 0.24 | 0.22 | 0.54 | |
| LSTM-MTLBackbone=BiLSTM-CNN-CRF2022.03 | 0.19 | 0.18 | 0.18 | 0.24 | 0.25 | 0.24 | 0.21 | — |