Research map / Label-efficient and robust learning / Knowledge distillation
ProKD: An Unsupervised Prototypical Knowledge Distillation Network for Zero-Resource Cross-Lingual Named Entity Recognition
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 ProKD instead of RIKD | Zero-resource cross-lingual NER · CoNLL-2002/2003 average | F1: 78.59 -> 80.35; 1.76% | Helped |
| Use ProKD instead of RIKD | Zero-resource cross-lingual NER · Wikiann average | F1: 55.55 -> 57.81; 2.26% | Helped |
| Add prototypical class-wise alignment (ablation) | ProKD · CoNLL-2002/2003 average | F1: 78.25 -> 80.35; ↓2.10 | Helped |
| Add prototypical self-training (ablation) | ProKD · CoNLL-2002/2003 average | F1: 78.80 -> 80.35; ↓1.55 | Helped |
| Add teacher prior knowledge to self-training (ablation) | ProKD · CoNLL-2002/2003 average | F1: 79.97 -> 80.35; ↓0.38 | Helped |
| Use cumulative learning instead of fixed α = 0.5 (ablation) | ProKD · CoNLL-2002/2003 average | F1: 80.08 -> 80.35; ↓0.27 | Helped |
| Add prototypical class-wise alignment (ablation) | ProKD · Wikiann average | F1: 55.82 -> 57.81; ↓1.99 | Helped |
| Add prototypical self-training (ablation) | ProKD · Wikiann average | F1: 56.43 -> 57.81; ↓1.38 | Helped |
| Add teacher prior knowledge to self-training (ablation) | ProKD · Wikiann average | F1: 57.40 -> 57.81; ↓0.41 | Helped |
| Use cumulative learning instead of fixed α = 0.5 (ablation) | ProKD · Wikiann average | F1: 57.65 -> 57.81; ↓0.16 | Helped |
Closest papers by meaning
- MCL-NER: Cross-Lingual Named Entity Recognition via Multi-View Contrastive Learning (AAAI 2024 · similarity 0.90)
- Discrepancy and Uncertainty Aware Denoising Knowledge Distillation for Zero-Shot Cross-Lingual Named Entity Recognition (AAAI 2024 · similarity 0.90)
- UniversalNER: Targeted Distillation from Large Language Models for Open Named Entity Recognition (ICLR 2024 · similarity 0.90)
- Three Heads Are Better than One: Improving Cross-Domain NER with Progressive Decomposed Network (AAAI 2024 · similarity 0.88)
- ConsistNER: Towards Instructive NER Demonstrations for LLMs with the Consistency of Ontology and Context (AAAI 2024 · similarity 0.87)
- HLMEA: Unsupervised Entity Alignment Based on Hybrid Language Models (AAAI 2025 · 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.