Research map / Architectures and efficiency / RNNs and normalization
A Context-Enriched Neural Network Method for Recognizing Lexical Entailment
Kun Zhang, Enhong Chen, Qi Liu, Chuanren Liu, Guangyi Lv
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add gate unit (attention over contexts) to sCENN (ablation) | CENN · Kotlerman2010 | F1: 64.1 -> 65.7 | Improved |
| add gate unit to sCENN (ablation) | CENN · Bless2011 | F1: 93.0 -> 94.3 | Improved |
| add gate unit to sCENN (ablation) | CENN · Baroni2012 | F1: 71.3 -> 78.6 | Improved |
| add gate unit to sCENN (ablation) | CENN · Turney2014 | F1: 66.8 -> 69.4 | Improved |
| add gate unit to sCENN (ablation) | CENN · Levy2014 | F1: 88.7 -> 89.6 | Improved |
| add concat, difference and binding layer plus gate to sNN (ablation) | CENN · Turney2014 | F1: 41.6 -> 66.8 (sCENN) -> 69.4 (CENN) | Improved |
| add multiple contextual embeddings with attention (CENN) vs best single-vector SVM | RLE · Turney2014 | F1: 69.6 -> 69.4 | Hurt |
| CENN vs best baseline | RLE · Baroni2012 | F1: 77.4 -> 78.6 | Improved |
| restrict to only dominant contexts (ablation) | CENN · Bless2011 | F1: 94.3 -> 93.7% | Hurt |
| restrict to only dominant context (ablation) | CENN · Levy2014 | F1: 89.6 -> 88.7% | Hurt |
Closest papers by meaning
- Have We Solved The Hard Problem? It’s Not Easy! Contextual Lexical Contrast as a Means to Probe Neural Coherence (AAAI 2021 · similarity 0.86)
- SensEmBERT: Context-Enhanced Sense Embeddings for Multilingual Word Sense Disambiguation (AAAI 2020 · similarity 0.85)
- Exploring Knowledge Graphs in an Interpretable Composite Approach for Text Entailment (AAAI 2019 · similarity 0.85)
- Can Neural Networks Understand Logical Entailment? (ICLR 2018 · similarity 0.85)
- Infusing Knowledge into the Textual Entailment Task Using Graph Convolutional Networks (AAAI 2020 · similarity 0.85)
- Deep contextualized word representations (ICLR 2018 · similarity 0.85)
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.
