Research map / Learning theory and algorithms / Combinatorial and search optimization
Chinese NER with Height-Limited Constituent Parsing
Rui Wang, Xin Xin, Wei Chang, Kun Ming, Biao Li, Xin Fan
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 height-limited constituent parsing to joint POS+CWS semi-CRF (ablation) | semi-CRF with attention + POS + CWS · OntoNotes 4.0 test | F1: 75.67 -> 76.67 | Improved |
| add height-limited constituent parsing (ablation) | semi-CRF with POS+CWS · OntoNotes 4.0 dev | F1: 74.98 -> 75.37 | Improved |
| add height-limited parsing for OOV entities (ablation) | semi-CRF with POS+CWS · OntoNotes 4.0 test | OOV recall: 64.70 -> 66.38 | Improved |
| add POS tagging and word segmentation to AtSemiCRF (ablation) | AtSemiCRF · OntoNotes 4.0 test | F1: 75.42 -> 75.67 | Improved |
| replace LSTM span feature with attention-based span feature in semi-CRF (ablation) | SRSemiCRF · OntoNotes 4.0 test | F1: 73.78 -> 75.42 | Improved |
| use subtraction LSTM span features instead of SR span features (ablation) | SRSemiCRF · OntoNotes 4.0 test | F1: 73.78 -> 74.77 | Improved |
| semi-CRF vs BIO-CRF with lattice LSTM (ablation) | Zhang18Latt · OntoNotes 4.0 test | F1: 73.88 -> 75.42 | Improved |
| joint model on OOV entities vs previous methods (ablation) | Zhang18Latt · OntoNotes 4.0 test | OOV recall: 60.04 -> 66.38 | Improved |
| add parsing to semi-CRF for word segmentation (ablation) | SemiCRF · OntoNotes 4.0 word segmentation | F1: 95.30 -> 95.45 | Improved |
| add parsing to semi-CRF for POS tagging (ablation) | SemiCRF · OntoNotes 4.0 POS tagging | F1: 83.99 -> 84.40 | Improved |
| unified model vs SemiCRF on long entities (length above 7) (ablation) | SemiCRF · OntoNotes 4.0 test | F1: 70.26 -> 74.42 | Improved |
| unified model vs semi-CRF+POS+CWS in decoding time (ablation) | SemiCRF+POS+CWS · test set 4636 sentences | avg time per sentence: 0.70s -> 1.00s | Hurt |
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 0.86)
- ConsistNER: Towards Instructive NER Demonstrations for LLMs with the Consistency of Ontology and Context (AAAI 2024 · similarity 0.86)
- PaDeLLM-NER: Parallel Decoding in Large Language Models for Named Entity Recognition (NeurIPS 2024 · similarity 0.86)
- Small Language Model Makes an Effective Long Text Extractor (AAAI 2025 · similarity 0.86)
- Three Heads Are Better than One: Improving Cross-Domain NER with Progressive Decomposed Network (AAAI 2024 · similarity 0.86)
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.
