AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

A Neural Network Approach to Verb Phrase Ellipsis Resolution

Wei-Nan Zhang, Yue Zhang, Yuanxing Liu, Donglin Di, Ting Liu

AAAI 2019 · Architectures and efficiency › RNNs and normalization · approach: Non-autoregressive translation

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 hand-crafted features F of Kenyon-Dean et al. (ablation)Transformer · VPE detection 5-fold CVF1: 0.88 -> 0.96Improved
add features F (ablation)MLP · VPE detection 5-fold CVF1: 0.87 -> 0.95Improved
add features F (ablation)SVM · VPE detection 5-fold CVF1: 0.87 -> 0.90Improved
replace SVM/ML with Transformer, same featuresSVM · VPE detection 5-fold CVF1: 0.87 -> 0.88Improved
replace SVM+F with Transformer+FSVM+F · VPE detection 5-fold CVF1: 0.90 -> 0.96Improved
add features F to AttGRU encoder (ablation)MLP · VPE resolution 5-fold CVaccuracy: 0.53 -> 0.87Improved
add features F to GRU encoder (ablation)MLP · VPE resolution 5-fold CVaccuracy: 0.51 -> 0.78Improved
add features F to Sum encoder (ablation)Transformer · VPE resolution 5-fold CVaccuracy: 0.83 -> 0.88Improved
use attention-based GRU instead of sum pooling (ablation)MLP+F · VPE resolution 5-fold CVaccuracy: 0.76 -> 0.87Improved
use attention-based GRU instead of sum pooling (ablation)Transformer+F · VPE resolution 5-fold CVaccuracy: 0.88 -> 0.94Improved
use Transformer instead of MLP without features, GRU encoder (ablation)GRU encoder · VPE resolution 5-fold CVaccuracy: 0.51 -> 0.85Improved
neural models vs MIRA baseline in resolutionAttGRU+F · VPE resolution 5-fold CVaccuracy: 0.65 -> 0.94Improved

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