Research map / Architectures and efficiency / Attention and transformers
Entity Structure Within and Throughout: Modeling Mention Dependencies for Document-Level 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 structured self-attention with biaffine transformation | RoBERTa Large on DocRED · DocRED test | F1: 60.54 -> 61.42 | Helped |
| Pretrain SSAN on distantly supervised data before annotated-data finetuning | SSAN Biaffine (RoBERTa Large) · DocRED test | F1: 65.92 | Helped |
| Add structured self-attention with biaffine transformation | SciBERT on CDR · CDR test | F1: 65.8 -> 68.7 | Helped |
| Add structured self-attention with biaffine transformation | BERT Large on GDA · GDA test | F1: 81.6 -> 83.9 | Helped |
| Add intra+coref dependency (ablation) | SSAN Biaffine (RoBERTa Large) · DocRED dev | F1: 61.57 -> 62.08 | Helped |
| Add intra+relate dependency (ablation) | SSAN Biaffine (RoBERTa Large) · DocRED dev | F1: 61.91 -> 62.08 | Helped |
| Add inter+coref dependency (ablation) | SSAN Biaffine (RoBERTa Large) · DocRED dev | F1: 61.74 -> 62.08 | Helped |
| Add inter+relate dependency (ablation) | SSAN Biaffine (RoBERTa Large) · DocRED dev | F1: 61.84 -> 62.08 | Helped |
| Add intraNE dependency (ablation) | SSAN Biaffine (RoBERTa Large) · DocRED dev | F1: 61.97 -> 62.08 | Helped |
| Add all five entity dependencies (ablation) | RoBERTa Large baseline · DocRED dev | F1: 60.58 -> 62.08 | Helped |
| Add dependency-specific prior bias (ablation) | RoBERTa Large baseline · DocRED dev | F1: 60.58 -> 60.59 | No clear effect |
| Add query-conditioned bias (ablation) | RoBERTa Large baseline · DocRED dev | F1: 60.58 -> 61.31 | Helped |
Closest papers by meaning
- Revisiting Document-Level Relation Extraction with Context-Guided Link Prediction (AAAI 2024 · similarity 0.90)
- A Hierarchical Network for Multimodal Document-Level Relation Extraction (AAAI 2024 · similarity 0.89)
- DocKS-RAG: Optimizing Document-Level Relation Extraction through LLM-Enhanced Hybrid Prompt Tuning (ICML 2025 · similarity 0.89)
- An Autoregressive Text-to-Graph Framework for Joint Entity and Relation Extraction (AAAI 2024 · similarity 0.88)
- COMM: Concentrated Margin Maximization for Robust Document-Level Relation Extraction (AAAI 2025 · similarity 0.87)
- A Positive-Unlabeled Metric Learning Framework for Document-Level Relation Extraction with Incomplete Labeling (AAAI 2024 · similarity 0.87)
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.