AI Paper Atlas

Research map / Label-efficient and robust learning / Knowledge distillation

Distantly-Supervised Named Entity Recognition with Adaptive Teacher Learning and Fine-Grained Student Ensemble

Xiaoye Qu, Jun Zeng, Daizong Liu, Zhefeng Wang, Baoxing Huai, Pan Zhou

AAAI 2023 · Label-efficient and robust learning › Knowledge distillation · approach: Few-shot and metric distillation · 15 citations (OpenAlex)

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 changedWhereMeasuredEffect
use ATSEN instead of SCDLDS-NER · CoNLL03F1: 83.69 -> 85.59Improved
use ATSEN instead of SCDLDS-NER · OntoNotes 5.0F1: 68.61 -> 68.95Improved
use ATSEN instead of SCDLDS-NER · WebpageF1: 68.47 -> 70.55Improved
use ATSEN instead of SCDLDS-NER · TwitterF1: 51.09 -> 52.46Improved
add consistent prediction (CP) (ablation)ATSEN · CoNLL03F1: 82.56 -> 85.59Improved
add threshold prediction (TP) (ablation)ATSEN · CoNLL03F1: 83.54 -> 85.59Improved
add cross-entropy loss (CE) (ablation)ATSEN · CoNLL03F1: 85.20 -> 85.59Improved
add adaptive distillation (AD) (ablation)ATSEN · CoNLL03F1: 84.95 -> 85.59Improved
add fine-grained ensemble (FE) (ablation)ATSEN · CoNLL03F1: 84.11 -> 85.59Improved
use adaptive rather than average distillation (ablation)ATSEN student training · CoNLL03F1: 85.19 -> 85.59Improved
use adaptive rather than manually weighted distillation (ablation)ATSEN student training · CoNLL03F1: 85.22 -> 85.59Improved
use adaptive rather than dynamically weighted distillation (ablation)ATSEN student training · CoNLL03F1: 85.26 -> 85.59Improved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy