Research map / Graph learning / Knowledge and temporal graphs
Learning Term Embeddings for Lexical Taxonomies
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add term-level attention to the hyponym-hypernym objective (ablation) | LTE · Pro5K | accuracy (%): 85.3 -> 86.7 | Helped |
| Add sibling-relation modeling (ablation) | LTE · Pro5K | accuracy (%): 86.7 -> 89.1 | Helped |
| Train ranked samples with curriculum learning (ablation) | LTE with sibling and hyponym-hypernym objectives · Pro5K | accuracy (%): 89.1 -> 89.9 | Helped |
| Add term-level attention to the hyponym-hypernym objective (ablation) | LTE · Pro300K | accuracy (%): 84.4 -> 85.6 | Helped |
| Add sibling-relation modeling (ablation) | LTE · Pro300K | accuracy (%): 85.6 -> 88.4 | Helped |
| Train ranked samples with curriculum learning (ablation) | LTE with sibling and hyponym-hypernym objectives · Pro300K | accuracy (%): 88.4 -> 89.2 | Helped |
| Add Pro5K term embeddings to BERT article vectors | BERTBASE · AG's News | accuracy (%): 92.0 -> 93.4 | Helped |
| Add Pro5K term embeddings to RoBERTa article vectors | RoBERTaBASE · AG's News | accuracy (%): 93.4 -> 94.1 | Helped |
| Add RotatE term embeddings to RoBERTa article vectors | RoBERTaBASE · AG's News | accuracy (%): 93.4 -> 94.1 | Helped |
| Add DistMult term embeddings to RoBERTa article vectors | RoBERTaBASE · AG's News | accuracy (%): 93.4 -> 94.1 | Helped |
Closest papers by meaning
- ExTaSem! Extending, Taxonomizing and Semantifying Domain Terminologies (AAAI 2016 · similarity 0.89)
- Semantic Structure-Based Word Embedding by Incorporating Concept Convergence and Word Divergence (AAAI 2018 · similarity 0.89)
- Unseen Word Representation by Aligning Heterogeneous Lexical Semantic Spaces (AAAI 2019 · similarity 0.88)
- KEML: A Knowledge-Enriched Meta-Learning Framework for Lexical Relation Classification (AAAI 2021 · similarity 0.87)
- Interpreting Knowledge Graph Relation Representation from Word Embeddings (ICLR 2021 · similarity 0.87)
- Antonym-Synonym Classification Based on New Sub-Space Embeddings (AAAI 2019 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.