Research map / Graph learning / Knowledge and temporal graphs
LADA-Trans-NER: Adaptive Efficient Transformer for Chinese Named Entity Recognition Using Lexicon-Attention and Data-Augmentation
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 lexicon-attention instead of SoftLexicon (ablation) | LADA-Trans-NER · Resume, MSRA, Weibo, OntoNotes, CTI | precision: 3.41%↓ on Resume, 5.81%↓ on MSRA, 13.86%↓ on Weibo, 7.81%↓ on OntoNotes, 1.95%↓ on CTI | Helped |
| Add AR-GAN data augmentation (ablation) | LADA-Trans-NER · Resume, MSRA, Weibo, OntoNotes, CTI | F1: 3.47%↓ on Resume, 1.70%↓ on MSRA, 2.79%↓ on Weibo, 0.99%↓ on OntoNotes; 3.59% on CTI | Helped |
| Use Transformer instead of CNN for sequence encoding (ablation) | LADA-Trans-NER · Resume, MSRA, Weibo, OntoNotes, CTI | F1: 98.70 vs 91.86 on Resume; 98.50 vs 91.85 on MSRA; 70.18 vs 54.10 on Weibo; 85.91 vs 75.85 on OntoNotes; 95.23 vs 89.01 o | Helped |
| Use Transformer instead of LSTM for sequence encoding (ablation) | LADA-Trans-NER · Resume, MSRA, Weibo, OntoNotes, CTI | F1: 98.70 vs 97.12 on Resume; 98.50 vs 98.89 on MSRA; 70.18 vs 68.10 on Weibo; 85.91 vs 85.36 on OntoNotes; 95.23 vs 93.73 o | Mixed |
| Add dropout (ablation) | LADA-Trans-NER · Resume, MSRA, Weibo, OntoNotes, CTI | F1: 96.15 -> 98.70; 96.12 -> 98.50; 68.53 -> 70.18; 84.89 -> 85.91; 90.45 -> 95.23 | Helped |
| Use word embedding dimension 50 rather than 10, 30, 100, 150, or 200 (ablation) | LADA-Trans-NER · Resume, MSRA, Weibo, OntoNotes, CTI | F1: Resume: 92.13, 94.82, 98.70, 97.12, 96.95, 95.23; MSRA: 92.83, 94.36, 98.50, 97.01, 96.63, 95.52; Weibo: 65.59, 66.63, 7 | Helped |
| Use learning rate decay (ablation) | LADA-Trans-NER · all evaluated datasets | accuracy | Helped |
Closest papers by meaning
- SENCR: A Span Enhanced Two-Stage Network with Counterfactual Rethinking for Chinese NER (AAAI 2024 · similarity 0.89)
- Span Graph Transformer for Document-Level Named Entity Recognition (AAAI 2024 · similarity 0.88)
- Small Language Model Makes an Effective Long Text Extractor (AAAI 2025 · similarity 0.86)
- PaDeLLM-NER: Parallel Decoding in Large Language Models for Named Entity Recognition (NeurIPS 2024 · similarity 0.85)
- Three Heads Are Better than One: Improving Cross-Domain NER with Progressive Decomposed Network (AAAI 2024 · similarity 0.85)
- MCL-NER: Cross-Lingual Named Entity Recognition via Multi-View Contrastive Learning (AAAI 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.