Research map / Graph learning / Knowledge and temporal graphs
GDPNet: Refining Latent Multi-View Graph for Relation Extraction
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 GDPNet's latent multi-view graph refinement to BERTs | BERTs · DialogRE test | F1: 61.2 -> 64.9 | Helped |
| Add GDPNet's latent multi-view graph refinement to BERTs | BERTs · DialogRE test | F1c: 55.4 -> 60.1 | Helped |
| Use GDPNet instead of SpanBERT | SpanBERT · TACRED | F1: 70.8 -> 70.5 | Hurt |
| Use GDPNet instead of SpanBERT | SpanBERT · TACRED-Revisit | F1: 78.0* -> 80.2 | Helped |
| Add multiple graph views instead of using a single homogeneous view (ablation) | GDPNet · DialogRE test | F1: 63.5 -> 64.9 | Helped |
| Use GGG instead of multi-head attention to initialize graph edges (ablation) | GDPNet · DialogRE test | F1: 62.1 -> 64.9 | Helped |
| Add the SoftDTW loss term (ablation) | GDPNet · DialogRE test | F1: 63.4 -> 64.9 | Helped |
| Add all DTWPool layers (ablation) | GDPNet · DialogRE test | F1: 48.9 -> 64.9 | Helped |
| Add all DTWPool layers (ablation) | GDPNet · DialogRE test | F1c: 22.4 -> 60.1 | Helped |
| Add GGG and all DTWPool layers (ablation) | GDPNet without GGG and DTWPool · DialogRE test | F1: 48.2 -> 64.9 | Helped |
Closest papers by meaning
- Multi-View Consistency for Relation Extraction via Mutual Information and Structure Prediction (AAAI 2020 · similarity 0.90)
- An Autoregressive Text-to-Graph Framework for Joint Entity and Relation Extraction (AAAI 2024 · similarity 0.88)
- Reliable Data Generation and Selection for Low-Resource Relation Extraction (AAAI 2024 · similarity 0.86)
- DocKS-RAG: Optimizing Document-Level Relation Extraction through LLM-Enhanced Hybrid Prompt Tuning (ICML 2025 · similarity 0.86)
- A Hierarchical Network for Multimodal Document-Level Relation Extraction (AAAI 2024 · similarity 0.86)
- Enhancing Low-Resource Relation Representations through Multi-View Decoupling (AAAI 2024 · similarity 0.86)
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.