Wei Wu
Research Subjects
Research Interests
Wei Wu has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Bin Shi
Xi'an Jiaotong University, China
1 shared publication - Bo Dong
School of Distance Education, Xi'an Jiaotong University, Xi'an, China
1 shared publication - Guangtao Wang
Senior Machine Learning Engineer, TikTok, San Jose, CA, USA
1 shared publication - Qinghua Zheng
School of Computer Science and Technology, Xian Jiaotong University, Xian, China
1 shared publication - Yanping Chen
State Key Laboratory of Public Big Data, Guizhou University, Guiyang, China
1 shared publication - Yuefei Wu
Xi'an Jiaotong University, Xi'an, China
1 shared publication
Latest Publications
- 11 July 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Audio, Speech, and Language Processing
- Vol. 31 (23299290) , 2899-2909
Exploring Interactive and Contrastive Relations for Nested Named Entity Recognition
Nested named entities (nested NEs) refer to the situation where one named entity is included or nested within another named entity, which cannot be recognized by the traditional sequence labeling methods. Recently, span-based methods have become the mainstream methods for nested Named Entity Recognition (nested NER). The fundamental concept behind this method is to enumerate nearly all potential spans as entity mentions and subsequently classify them. However, span-based methods independently classify spans without considering the semantic relations among them, which negatively impacts the span representation. To address the issue, we propose a novel deep learning architecture for nested NER that explores interactive and contrastive relations among spans. Specifically, we design a scale transformation mechanism that embeds geometric information into span representations, which enhances the model's ability to encode interactive relations between spans. Additionally, we introduce a supervised contrastive learning loss that pulls apart highly overlapping spans in the embedding space to encode the contrastive relations. Experiments show that our method achieves state-of-the-art or competitive performance on three publicly nested NER datasets, thus validating its effectiveness.
- 11 July 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Audio, Speech, and Language Processing
- Vol. 31 (23299290) , 2899-2909
Exploring Interactive and Contrastive Relations for Nested Named Entity Recognition
Nested named entities (nested NEs) refer to the situation where one named entity is included or nested within another named entity, which cannot be recognized by the traditional sequence labeling methods. Recently, span-based methods have become the mainstream methods for nested Named Entity Recognition (nested NER). The fundamental concept behind this method is to enumerate nearly all potential spans as entity mentions and subsequently classify them. However, span-based methods independently classify spans without considering the semantic relations among them, which negatively impacts the span representation. To address the issue, we propose a novel deep learning architecture for nested NER that explores interactive and contrastive relations among spans. Specifically, we design a scale transformation mechanism that embeds geometric information into span representations, which enhances the model's ability to encode interactive relations between spans. Additionally, we introduce a supervised contrastive learning loss that pulls apart highly overlapping spans in the embedding space to encode the contrastive relations. Experiments show that our method achieves state-of-the-art or competitive performance on three publicly nested NER datasets, thus validating its effectiveness.