Research map / Label-efficient and robust learning / Recognition and long-tail
Multi-Labeled Relation Extraction with Attentive Capsule Network
Xinsong Zhang, Pengshuai Li, Weijia Jia, Hai Zhao
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 |
|---|---|---|---|
| use attentive capsule routing instead of max-pooling or word-level aggregation | RNN relation extraction · NYT-10 | F1: 38.4 -> 41.6 | Improved |
| use attentive capsule routing instead of max-pooling or word-level aggregation | RNN relation extraction · SemEval-2010 Task 8 | F1: 84.0 -> 84.5 | Improved |
| use attentive capsule routing instead of max-pooling | CNN relation extraction · NYT-10 | F1: 37.8 -> 38.8 | Improved |
| use attentive capsule routing instead of max-pooling | CNN relation extraction · SemEval-2010 Task 8 | F1: 69.7 -> 80.4 | Improved |
| use the RNN-based rather than CNN-based attentive capsule model | Att-CapNet · SemEval-2010 Task 8 | F1: 80.4 -> 84.5 | Improved |
| use attentive capsule clustering rather than max-pooling RNN | RNN relation extraction · NYT-10 selected 500 multi-labeled sentences | F1: 90.5 -> 91.8 | Improved |
| add attention-based routing (ablation) | Att-CapNet (RNN-based) · SemEval-2010 Task 8 | F1: 83.6 -> 84.5 | Improved |
| add sliding-margin loss instead of fixed-margin loss (ablation) | Att-CapNet (RNN-based) · SemEval-2010 Task 8 | F1: 82.3 -> 84.5 | Improved |
Closest papers by meaning
- Aggregating Inter-Sentence Information to Enhance Relation Extraction (AAAI 2016 · similarity 0.88)
- Joint Extraction of Entities and Overlapping Relations Using Position-Attentive Sequence Labeling (AAAI 2019 · similarity 0.87)
- Are Noisy Sentences Useless for Distant Supervised Relation Extraction? (AAAI 2020 · similarity 0.87)
- Enhancing Low-Resource Relation Representations through Multi-View Decoupling (AAAI 2024 · similarity 0.86)
- A Positive-Unlabeled Metric Learning Framework for Document-Level Relation Extraction with Incomplete Labeling (AAAI 2024 · similarity 0.85)
- A Hierarchical Network for Multimodal Document-Level Relation Extraction (AAAI 2024 · similarity 0.84)
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.
