Research map / Optimization / Optimizers
DeepType: Multilingual Entity Linking by Neural Type System Evolution
Jonathan Raiman, Olivier Raiman
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 an automatically discovered type system to LinkCount | DeepType · Wiki-Disamb30 | accuracy: 82.710 -> 92.375 | Improved |
| Add an automatically discovered type system to LinkCount | DeepType · CoNLL (YAGO) | accuracy: 68.614 -> 94.879 | Improved |
| Add an automatically discovered type system to LinkCount | DeepType · TAC KBP 2010 | accuracy: 81.485 -> 90.850 | Improved |
| Use DeepType-discovered types rather than the human-designed type system | DeepType entity linking · Wiki-Disamb30 | accuracy: 91.888 -> 92.375 | Improved |
| Use stochastic GA search rather than beam search for type discovery | DeepType type-system discovery · Type system discovery | accuracy: 97.84 -> 96.959 | Hurt |
| Use stochastic GA search rather than beam search for type discovery | DeepType type-system discovery · Type system discovery | function evaluations: 5.12\cdot 10^7 -> 116,000 | Improved |
| Add DeepType pretraining to Bi-LSTM-CNN (ablation) | NER model · CoNLL 2003 test | F1: 84.74 -> 88.67 | Improved |
| Add DeepType pretraining to Bi-LSTM-CNN (ablation) | NER model · OntoNotes dev | F1: 83.17 -> 85.11 | Improved |
| Add DeepType pretraining to Bi-LSTM-CNN (ablation) | NER model · CoNLL 2003 test and OntoNotes dev | F1: 84.74 -> 88.67; 83.17 -> 85.11 | Mixed |
| Train GA with English only instead of English and French (ablation) | DeepType GA · English Wikipedia and Wiki-Disamb30 | accuracy: 93.684 -> 93.029; 92.062 -> 91.743 | Hurt |
| Train GA with English only instead of English and French (ablation) | DeepType GA · CoNLL (YAGO) | accuracy: 94.879 -> 93.701 | Hurt |
Closest papers by meaning
- Seed-Guided Fine-Grained Entity Typing in Science and Engineering Domains (AAAI 2024 · similarity 0.85)
- End-to-End Ontology Learning with Large Language Models (NeurIPS 2024 · similarity 0.85)
- UniversalNER: Targeted Distillation from Large Language Models for Open Named Entity Recognition (ICLR 2024 · similarity 0.85)
- Entity Alignment with Noisy Annotations from Large Language Models (NeurIPS 2024 · similarity 0.85)
- HLMEA: Unsupervised Entity Alignment Based on Hybrid Language Models (AAAI 2025 · similarity 0.84)
- MCL-NER: Cross-Lingual Named Entity Recognition via Multi-View Contrastive Learning (AAAI 2024 · similarity 0.84)
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.
