Research map / Label-efficient and robust learning / Recognition and long-tail
Fine-Grained Entity Typing for Domain Independent Entity Linking
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 fine-grained entity typing with Wikipedia category supervision instead of direct entity-linking training | ET4EL · Unseen-Mentions test | accuracy: 62.2 vs 43.4 | Helped |
| Use fine-grained entity typing with Wikipedia category supervision instead of the strongest transductive CBoW baseline | ET4EL · CoNLL-YAGO testa (dev) | accuracy: 84.6 -> 88.1 | Helped |
| Add the first document sentence as context (ablation) | ET4EL · CoNLL-YAGO testa (dev) | accuracy: 83.8 -> 88.1 | Helped |
| Add left and right 50 tokens around the mention (ablation) | ET4EL · CoNLL-YAGO testa (dev) | accuracy: 83.8 -> 84.5 | Helped |
| Increase the category vocabulary from 1k to 60k categories (ablation) | ET4EL · CoNLL-YAGO testa (dev) | accuracy: 85.1 -> 88.1 | Helped |
| Expand original Wikipedia categories by splitting categories into useful parts (ablation) | ET4EL · CoNLL-YAGO testa (dev) | accuracy: 84.4 -> 87.1 | Helped |
| Prefer summed type probabilities over averaging or a log odds ratio (ablation) | ET4EL candidate scoring · Entity linking | accuracy | Helped |
Closest papers by meaning
- Improving Neural Fine-Grained Entity Typing With Knowledge Attention (AAAI 2018 · similarity 0.90)
- Fine-Grained Named Entity Typing over Distantly Supervised Data Based on Refined Representations (AAAI 2020 · similarity 0.90)
- Improving Entity Linking by Modeling Latent Entity Type Information (AAAI 2020 · similarity 0.89)
- Seed-Guided Fine-Grained Entity Typing in Science and Engineering Domains (AAAI 2024 · similarity 0.87)
- Three Heads Are Better than One: Improving Cross-Domain NER with Progressive Decomposed Network (AAAI 2024 · similarity 0.87)
- A Dual-Way Enhanced Framework from Text Matching Point of View for Multimodal Entity Linking (AAAI 2024 · 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.