Research map / Label-efficient and robust learning / Recognition and long-tail
Efficient Dependency-Guided Named Entity Recognition
Zhanming Jie, Aldrian Muis, Wei Lu
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 |
|---|---|---|---|
| Use dependency-guided valid spans instead of unrestricted semi-Markov spans | DGM versus Semi-Markov CRFs · OntoNotes 5.0 Broadcast News, given dependencies; dependency features | F-score: 79.6 -> 80.5 | Improved |
| Use dependency-guided valid spans instead of unrestricted semi-Markov spans | DGM versus Semi-Markov CRFs · OntoNotes 5.0 Broadcast News, predicted dependencies; dependency features | F-score: 79.5 -> 79.4 | No clear effect |
| Allow valid spans covered by multiple adjacent dependency arcs, rather than only single arcs (ablation) | DGM · OntoNotes 5.0 Broadcast News, given dependencies; dependency features | F-score: 79.0 -> 80.5 | Improved |
| Allow valid spans covered by multiple adjacent dependency arcs, rather than only single arcs (ablation) | DGM · OntoNotes 5.0 Broadcast News, predicted dependencies; dependency features | F-score: 78.2 -> 79.4 | Improved |
| Allow valid spans covered by multiple adjacent dependency arcs, rather than only single arcs (ablation) | DGM · OntoNotes 5.0 Broadcast News, given dependencies; no dependency features | F-score: 78.6 -> 79.9 | Improved |
| Allow valid spans covered by multiple adjacent dependency arcs, rather than only single arcs (ablation) | DGM · OntoNotes 5.0 Broadcast News, predicted dependencies; no dependency features | F-score: 78.0 -> 78.8 | Improved |
| Add dependency features (ablation) | DGM · OntoNotes 5.0 Broadcast News, given dependencies | F-score: 79.9 -> 80.5 | Improved |
| Add dependency features (ablation) | DGM · OntoNotes 5.0 Broadcast News, predicted dependencies | F-score: 78.8 -> 79.4 | Improved |
| Use predicted rather than given dependency trees | DGM · OntoNotes 5.0 Broadcast News; dependency features | F-score: 80.5 -> 79.4 | Hurt |
| Use predicted rather than given dependency trees | DGM · OntoNotes 5.0 Broadcast News; no dependency features | F-score: 79.9 -> 78.8 | Hurt |
| Use DGM rather than Semi-Markov CRFs | OntoNotes 5.0 Broadcast News · Training time per iteration | relative training time: 47% | Improved |
Closest papers by meaning
- Span Graph Transformer for Document-Level Named Entity Recognition (AAAI 2024 · similarity 0.88)
- ConsistNER: Towards Instructive NER Demonstrations for LLMs with the Consistency of Ontology and Context (AAAI 2024 · similarity 0.87)
- Small Language Model Makes an Effective Long Text Extractor (AAAI 2025 · similarity 0.86)
- UniversalNER: Targeted Distillation from Large Language Models for Open Named Entity Recognition (ICLR 2024 · similarity 0.86)
- MCL-NER: Cross-Lingual Named Entity Recognition via Multi-View Contrastive Learning (AAAI 2024 · similarity 0.85)
- GuideNER: Annotation Guidelines Are Better than Examples for In-Context Named Entity Recognition (AAAI 2025 · 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.
