Hierarchically-Refined Label Attention Network for Sequence Labeling

IJCNLP 2019 Leyang CuiYue Zhang

CRF has been used as a powerful model for statistical sequence labeling. For neural sequence labeling, however, BiLSTM-CRF does not always lead to better results compared with BiLSTM-softmax local classification... (read more)

PDF Abstract

Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT LEADERBOARD
CCG Supertagging CCGBank BiLSTM-LAN Accuracy 94.7 # 2
Named Entity Recognition Ontonotes v5 (English) BiLSTM-LAN F1 88.16 # 10
Part-Of-Speech Tagging Penn Treebank BiLSTM-LAN Accuracy 97.65 # 6
Part-Of-Speech Tagging UD BiLSTM-LAN Avg accuracy 96.88 # 1