AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

GRN: Gated Relation Network to Enhance Convolutional Neural Network for Named Entity Recognition

Hui Chen, Zijia Lin, Guiguang Ding, Jianguang Lou, Yusen Zhang, Borje Karlsson

AAAI 2019 · Architectures and efficiency › RNNs and normalization · approach: Efficient recurrent architectures

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
Add multi-scale context branches (ablation)GRN · CoNLL-2003F1: 88.36 ± 0.21 -> 91.44 ± 0.16Improved
Use multi-scale context branches rather than only the kernel-size-3 branch (ablation)GRN · CoNLL-2003F1: 90.88 ± 0.22 -> 91.44 ± 0.16Improved
Add the relation layer (ablation)GRN · CoNLL-2003F1: 90.13 ± 0.28 -> 91.44 ± 0.16Improved
Use channel-wise gated relation fusion rather than direct averaging (ablation)GRN · CoNLL-2003F1: 90.72 ± 0.06 -> 91.44 ± 0.16Improved
Use channel-wise gating rather than gated attention (ablation)GRN · CoNLL-2003F1: 87.11 ± 0.25 -> 91.44 ± 0.16Improved
Add multi-scale context branches (ablation)GRN · OntoNotes 5.0F1: 82.21 ± 0.23 -> 87.67 ± 0.17Improved
Use multi-scale context branches rather than only the kernel-size-3 branch (ablation)GRN · OntoNotes 5.0F1: 86.66 ± 0.21 -> 87.67 ± 0.17Improved
Add the relation layer (ablation)GRN · OntoNotes 5.0F1: 85.87 ± 0.16 -> 87.67 ± 0.17Improved
Use channel-wise gated relation fusion rather than direct averaging (ablation)GRN · OntoNotes 5.0F1: 85.81 ± 0.14 -> 87.67 ± 0.17Improved
Use channel-wise gating rather than gated attention (ablation)GRN · OntoNotes 5.0F1: 79.83 ± 0.37 -> 87.67 ± 0.17Improved
Use GRN instead of CNN-BiLSTM-CRFNER system · CoNLL-2003training speedup: 1.16xImproved
Use GRN instead of CNN-BiLSTM-CRFNER system · CoNLL-2003test speedup: 1.19xImproved

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