TY - GEN
T1 - Context Tracking Network
T2 - 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2021
AU - Zhang, Yingxue
AU - Meng, Fandong
AU - Li, Peng
AU - Jian, Ping
AU - Zhou, Jie
N1 - Publisher Copyright:
© 2021 Association for Computational Linguistics.
PY - 2021
Y1 - 2021
N2 - Implicit discourse relation recognition (IDRR) aims to identify logical relations between two adjacent sentences in the discourse. Existing models fail to fully utilize the contextual information which plays an important role in interpreting each local sentence. In this paper, we thus propose a novel graph-based Context Tracking Network (CT-Net) to model the discourse context for IDRR. The CT-Net firstly converts the discourse into the paragraph association graph (PAG), where each sentence tracks their closely related context from the intricate discourse through different types of edges. Then, the CT-Net extracts contextual representation from the PAG through a specially designed cross-grained updating mechanism, which can effectively integrate both sentence-level and token-level contextual semantics. Experiments on PDTB 2.0 show that the CT-Net gains better performance than models that roughly model the context.
AB - Implicit discourse relation recognition (IDRR) aims to identify logical relations between two adjacent sentences in the discourse. Existing models fail to fully utilize the contextual information which plays an important role in interpreting each local sentence. In this paper, we thus propose a novel graph-based Context Tracking Network (CT-Net) to model the discourse context for IDRR. The CT-Net firstly converts the discourse into the paragraph association graph (PAG), where each sentence tracks their closely related context from the intricate discourse through different types of edges. Then, the CT-Net extracts contextual representation from the PAG through a specially designed cross-grained updating mechanism, which can effectively integrate both sentence-level and token-level contextual semantics. Experiments on PDTB 2.0 show that the CT-Net gains better performance than models that roughly model the context.
UR - https://www.scopus.com/pages/publications/85115854583
U2 - 10.18653/v1/2021.naacl-main.126
DO - 10.18653/v1/2021.naacl-main.126
M3 - Conference contribution
AN - SCOPUS:85115854583
T3 - NAACL-HLT 2021 - 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Conference
SP - 1592
EP - 1599
BT - NAACL-HLT 2021 - 2021 Conference of the North American Chapter of the Association for Computational Linguistics
PB - Association for Computational Linguistics (ACL)
Y2 - 6 June 2021 through 11 June 2021
ER -