N
Ning Ding
Researcher at Tsinghua University
Publications - 12
Citations - 225
Ning Ding is an academic researcher from Tsinghua University. The author has contributed to research in topics: Language model & Computer science. The author has an hindex of 6, co-authored 12 publications receiving 92 citations.
Papers
More filters
Proceedings ArticleDOI
Hierarchy-Aware Global Model for Hierarchical Text Classification.
Jie Zhou,Chunping Ma,Dingkun Long,Guangwei Xu,Ning Ding,Haoyu Zhang,Pengjun Xie,Gongshen Liu +7 more
TL;DR: A novel end-to-end hierarchy-aware global model (HiAGM) with two variants, both of which achieve significant and consistent improvements on three benchmark datasets.
Journal ArticleDOI
Modeling Relation Paths for Knowledge Graph Completion
TL;DR: A novel method, Type-aware Attentive Path Reasoning (TAPR), is proposed to complete the knowledge graph by simultaneously considering KG structural information, textual information, and type information and describes a type-level attention to select the most relevant type of given entity in a specific triple without any predefined rules or patterns.
Posted Content
PTR: Prompt Tuning with Rules for Text Classification.
TL;DR: The authors proposed a prompt tuning with rules (PTR) approach for many-class text classification, and applied logic rules to construct prompts with several sub-prompts to encode prior knowledge of each class into prompt tuning.
Posted Content
Knowledgeable Prompt-tuning: Incorporating Knowledge into Prompt Verbalizer for Text Classification
TL;DR: In this article, the authors focus on incorporating external knowledge into the verbalizer, forming a knowledgeable prompt-tuning (KPT) to improve and stabilize the performance of pre-trained language models with task-specific prompts.
Proceedings ArticleDOI
Coupling Distant Annotation and Adversarial Training for Cross-Domain Chinese Word Segmentation
TL;DR: The essence of “Chinese words” is rethink and an automatic distant annotation mechanism is designed, which does not need any supervision or pre-defined dictionaries on the target domain, which is significantly outperforming previous state-of-the-arts cross-domain CWS methods.