Research map / Graph learning / Graph self-supervised learning
Multi-Task Learning for Metaphor Detection with Graph Convolutional Neural Networks and Word Sense Disambiguation
Duong Le, My Thai, Thien Nguyen
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 BiLSTM layer (ablation) | MUL GCN · VUA VERB | F1: 67.6 -> 71.7 | Improved |
| add control module (ablation) | MUL GCN · VUA VERB | F1: 68.3 -> 71.7 | Improved |
| add GCN module over dependency trees (ablation) | MUL GCN · VUA VERB | F1: 67.0 -> 71.7 | Improved |
| use GCN instead of self-attention (ablation) | MUL GCN · VUA VERB | F1: 69.9 -> 71.7 | Improved |
| add WSD multi-task learning with two encoders vs single network (ablation) | MUL GCN · VUA VERB | F1: 68.9 -> 71.7 | Improved |
| two separate encoders with similarity loss vs WSD pretraining (ablation) | MUL GCN · VUA VERB | F1: 69.6 -> 71.7 | Improved |
| two separate encoders vs one shared encoder with alternating training (ablation) | MUL GCN · VUA VERB | F1: 69.0 -> 71.7 | Improved |
| use MSE rather than KL as representation dissimilarity (ablation) | MUL GCN · VUA VERB | F1: 69.7 -> 71.7 | Improved |
| use MSE rather than cosine as dissimilarity (ablation) | MUL GCN · VUA VERB | F1: 68.4 -> 71.7 | Improved |
| use MSE rather than margin loss as dissimilarity (ablation) | MUL GCN · VUA VERB | F1: 68.3 -> 71.7 | Improved |
| MUL GCN classification model vs RNN MHCA sequence labeling | MUL GCN · MOH-X | F1: 80.0 -> 79.6 | No clear effect |
| MUL GCN vs RNN MHCA | MUL GCN · VUA VERB | F1: 70.5 -> 71.7 | Improved |
Closest papers by meaning
- Graph Convolutional Networks for Text Classification (AAAI 2019 · similarity 0.85)
- End-to-End Convolutional Semantic Embeddings (CVPR 2018 · similarity 0.83)
- Attention-Driven Dynamic Graph Convolutional Network for Multi-Label Image Recognition (ECCV 2020 · similarity 0.82)
- Distilling Knowledge From Graph Convolutional Networks (CVPR 2020 · similarity 0.82)
- A Survey on Graph Neural Networks and Graph Transformers in Computer Vision: A Task-Oriented Perspective (TPAMI 2024 · similarity 0.82)
- ConCon-Chi: Concept-Context Chimera Benchmark for Personalized Vision-Language Tasks (CVPR 2024 · similarity 0.81)
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.
