Lune

EMNLP Findings2021

HyperExpan: Taxonomy Expansion with Hyperbolic Representation Learning

Mingyu Derek Ma, Muhao Chen, Te-Lin Wu, Nanyun Peng

Abstract

Taxonomies are valuable resources for many applications, but the limited coverage due to the expensive manual curation process hinders their general applicability. Prior works attempt to automatically expand existing taxonomies to improve their coverage by learning concept embeddings in Euclidean space, while taxonomies, inherently hierarchical, more naturally align with the geometric properties of a hyperbolic space. In this paper, we present HyperExpan, a taxonomy expansion algorithm that seeks to preserve the structure of a taxonomy in a more expressive hyperbolic embedding space and learn to represent concepts and their relations with a Hyperbolic Graph Neural Network (HGNN). Specifically, Hyper-Expan leverages position embeddings to exploit the structure of the existing taxonomies, and characterizes the concept profile information to support the inference on unseen concepts during training. Experiments show that our proposed HyperExpan outperforms baseline models with representation learning in a Euclidean feature space and achieves state-ofthe-art performance on the taxonomy expansion benchmarks.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

11 top-tier papers have cited it since. Know what they found before your reviewers do.

Questions to start from

Lune can draw on

the full paper · the top-tier papers that cite it · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Hyperbolic Interaction Model for Hierarchical Multi-Label Classification cites Hyperbolic Neural Networks++
  • DICE: Data-Efficient Clinical Event Extraction with Generative Models cites Can NLI Provide Proper Indirect Supervision for Low-resource Biomedical Relation Extraction?
  • Can NLI Provide Proper Indirect Supervision for Low-resource Biomedical Relation Extraction? cites DICE: Data-Efficient Clinical Event Extraction with Generative Models
  • TacoPrompt: A Collaborative Multi-Task Prompt Learning Method for Self-Supervised Taxonomy Completion cites A Single Vector Is Not Enough: Taxonomy Expansion via Box Embeddings
  • TacoPrompt: A Collaborative Multi-Task Prompt Learning Method for Self-Supervised Taxonomy Completion cites TaxoExpan: Self-supervised Taxonomy Expansion with Position-Enhanced Graph Neural Network
  • A Single Vector Is Not Enough: Taxonomy Expansion via Box Embeddings cites TaxoExpan: Self-supervised Taxonomy Expansion with Position-Enhanced Graph Neural Network
  • Constant Curvature Graph Convolutional Networks cites Hyperbolic Neural Networks++
  • Taxonomy Completion via Implicit Concept Insertion cites TaxoExpan: Self-supervised Taxonomy Expansion with Position-Enhanced Graph Neural Network
Builds onCited byLeft to right by year; the more cited, the higher.

Cited by top-tier papers11

  1. A Single Vector Is Not Enough: Taxonomy Expansion via Box EmbeddingsSong Jiang, Qiyue Yao, Qifan Wang, Yizhou SunWWW 2023 · 20 citations
  2. DICE: Data-Efficient Clinical Event Extraction with Generative ModelsMingyu Derek Ma, Alexander Taylor, Wei Wang, Nanyun PengACL 2023 · 18 citations
  3. Can NLI Provide Proper Indirect Supervision for Low-resource Biomedical Relation Extraction?Jiashu Xu, Mingyu Derek Ma, Muhao ChenACL 2023 · 14 citations
  4. Taxonomy Completion via Implicit Concept InsertionJingchuan Shi, Hang Dong, Jiaoyan Chen, Zhe Wu et al.WWW 2024 · 10 citations
  5. TacoPrompt: A Collaborative Multi-Task Prompt Learning Method for Self-Supervised Taxonomy CompletionHongyuan Xu, Ciyi Liu, Yuhang Niu, Yunong Chen et al.EMNLP 2023 · 9 citations

Builds on12

  1. Hyperbolic Neural Networks++Ryohei Shimizu, Yusuke Mukuta, Tatsuya HaradaICLR 2021 · 791 citations
  2. Learning from History: Modeling Temporal Knowledge Graphs with Sequential Copy-Generation NetworksCunchao Zhu, Muhao Chen, Changjun Fan, Guangquan Cheng et al.AAAI 2021 · 343 citations
  3. Constant Curvature Graph Convolutional NetworksGregor Bachmann, Gary Bécigneul, Octavian GaneaICML 2020 · 169 citations
  4. TaxoExpan: Self-supervised Taxonomy Expansion with Position-Enhanced Graph Neural NetworkJiaming Shen, Zhihong Shen, Chenyan Xiong, Chi Wang et al.WWW 2020 · 85 citations
  5. Hyperbolic Interaction Model for Hierarchical Multi-Label ClassificationBoli Chen, Xin Huang, Lin Xiao, Zixin Cai et al.AAAI 2020 · 78 citations

Related papers

  1. Knowledge Association with Hyperbolic Knowledge Graph EmbeddingsZequn Sun, Muhao Chen, Wei Hu, Chengming Wang et al.EMNLP 2020 · 72 citations
  2. Mixed-Curvature Multi-Relational Graph Neural Network for Knowledge Graph CompletionShen Wang, Xiaokai Wei, Cícero Nogueira dos Santos, Zhiguo Wang et al.WWW 2021 · 122 citations
  3. Bringing Motion Taxonomies to Continuous Domains via GPLVM on Hyperbolic manifoldsNoémie Jaquier, Leonel Rozo, Miguel González Duque, Viacheslav Borovitskiy et al.ICML 2024 · 11 citations
  4. Discrete-time Temporal Network Embedding via Implicit Hierarchical Learning in Hyperbolic SpaceMenglin Yang, Min Zhou, Marcus Kalander, Zengfeng Huang et al.KDD 2021 · 101 citations
  5. TEMP: Taxonomy Expansion with Dynamic Margin Loss through Taxonomy-PathsZichen Liu, Hongyuan Xu, Yanlong Wen, Ning Jiang et al.EMNLP 2021 · 16 citations
Dusk over the sea between two cliffs drawn in fine vertical lines