Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
KEML: A Knowledge-Enriched Meta-Learning Framework for Lexical Relation Classification
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 |
|---|---|---|---|
| use KEML instead of SphereRE | lexical relation classification · K&H+N | F1: 0.990 -> 0.993 | Helped |
| use KEML instead of SphereRE | lexical relation classification · BLESS | F1: 0.938 -> 0.944 | Helped |
| use KEML instead of SphereRE | lexical relation classification · ROOT09 | F1: 0.861 -> 0.878 | Helped |
| use KEML instead of SphereRE | lexical relation classification · EVALution | F1: 0.62 -> 0.660 | Helped |
| add meta-learning before supervised classification (ablation) | KEML · K&H+N | F1: 0.984 -> 0.993 | Helped |
| add meta-learning before supervised classification (ablation) | KEML · BLESS | F1: 0.941 -> 0.944 | Helped |
| add meta-learning before supervised classification (ablation) | KEML · ROOT09 | F1: 0.873 -> 0.878 | Helped |
| add meta-learning before supervised classification (ablation) | KEML · EVALution | F1: 0.644 -> 0.660 | Helped |
| add meta-learning before supervised classification (ablation) | KEML · CogALex-V Subtask 2 | F1 (All): 0.485 -> 0.500 | Helped |
| combine binary and multi-way knowledge-base objectives (ablation) | LKB-BERT · K&H+N validation | F1: 0.964 -> 0.983; 0.972 -> 0.983 | Helped |
| combine binary and multi-way knowledge-base objectives (ablation) | LKB-BERT · BLESS validation | F1: 0.921 -> 0.939; 0.929 -> 0.939 | Helped |
| combine binary and multi-way knowledge-base objectives (ablation) | LKB-BERT · ROOT09 validation | F1: 0.854 -> 0.863; 0.861 -> 0.863 | Helped |
Closest papers by meaning
- Learning Term Embeddings for Lexical Taxonomies (AAAI 2021 · similarity 0.87)
- Inducing Relational Knowledge from BERT (AAAI 2020 · similarity 0.86)
- DocKS-RAG: Optimizing Document-Level Relation Extraction through LLM-Enhanced Hybrid Prompt Tuning (ICML 2025 · similarity 0.86)
- Learning Concept Prerequisite Relation via Global Knowledge Relation Optimization (AAAI 2025 · similarity 0.85)
- Enhancing Low-Resource Relation Representations through Multi-View Decoupling (AAAI 2024 · similarity 0.85)
- Linearity of Relation Decoding in Transformer Language Models (ICLR 2024 · similarity 0.85)
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.