AI Paper Atlas

Research map / Graph learning / Graph algorithms and clustering

Strongly Incremental Constituency Parsing with Graph Neural Networks

Kaiyu Yang, Jia Deng

NeurIPS 2020 · Graph learning › Graph algorithms and clustering · approach: Neural combinatorial optimization on graphs · 22 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 attach-juxtapose instead of ISR (ablation)Parser with XLNet and GCN decoder · PTB testF1: 96.23 ± 0.04 -> 96.34 ± 0.03Improved
Use attach-juxtapose instead of ISR (ablation)Parser with XLNet and GCN decoder · PTB testEM: 58.99 ± 0.11 -> 59.17 ± 0.33Improved
Use a graph-based decoder rather than a sequence-based decoder (ablation)Parser with BERT · PTB testF1: 94.89 ± 0.03 -> 95.79 ± 0.05Improved
Use a graph-based decoder rather than a sequence-based decoder (ablation)Parser with BERT · PTB testEM: 53.26 ± 0.45 -> 57.29 ± 0.57Improved
Use a graph-based decoder rather than a sequence-based decoder (ablation)Parser with XLNet · PTB testF1: 95.54 ± 0.07 -> 96.34 ± 0.03Improved
Use a graph-based decoder rather than a sequence-based decoder (ablation)Parser with XLNet · PTB testEM: 55.84 ± 0.53 -> 59.17 ± 0.33Improved
Use a graph-based decoder rather than a sequence-based decoder (ablation)Parser with BERT · CTB testF1: 90.98 ± 0.35 -> 93.59 ± 0.26Improved
Use a graph-based decoder rather than a sequence-based decoder (ablation)Parser with BERT · CTB testEM: 44.14 ± 0.88 -> 49.72 ± 0.83Improved

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