Research map / Label-efficient and robust learning / Semi-supervised learning
Competition or Cooperation? Exploring Unlabeled Data via Challenging Minimax Game for Semi-supervised 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 |
|---|---|---|---|
| Use AdvSRE's minimax competition instead of GradLRE's gradient imitation | BERT-based semi-supervised relation extraction · SemEval, 5% labeled / 50% unlabeled | F1: 79.65 -> 81.90 | Helped |
| Use AdvSRE's minimax competition instead of GradLRE's gradient imitation | BERT-based semi-supervised relation extraction · SemEval, 10% labeled / 50% unlabeled | F1: 81.69 -> 82.72 | Helped |
| Use AdvSRE's minimax competition instead of GradLRE's gradient imitation | BERT-based semi-supervised relation extraction · SemEval, 30% labeled / 50% unlabeled | F1: 85.52 -> 86.42 | Helped |
| Use AdvSRE's minimax competition instead of GradLRE's gradient imitation | BERT-based semi-supervised relation extraction · TACRED, 3% labeled / 50% unlabeled | F1: 47.37 -> 48.38 | Helped |
| Use AdvSRE's minimax competition instead of GradLRE's gradient imitation | BERT-based semi-supervised relation extraction · TACRED, 10% labeled / 50% unlabeled | F1: 58.20 -> 61.10 | Helped |
| Use AdvSRE's minimax competition instead of GradLRE's gradient imitation | BERT-based semi-supervised relation extraction · TACRED, 15% labeled / 50% unlabeled | F1: 59.93 -> 62.53 | Helped |
| Pretrain the discriminator on labeled data before jointly training it with the generator (ablation) | AdvSRE · SemEval and TACRED, low-label settings | F1 | Hurt |
| Train the discriminator on labeled data separately from its competition with the generator (ablation) | AdvSRE · SemEval and TACRED, low-label settings | F1 | Hurt |
Closest papers by meaning
- Few-Shot, No Problem: Descriptive Continual Relation Extraction (AAAI 2025 · similarity 0.85)
- Synergistic Anchored Contrastive Pre-training for Few-Shot Relation Extraction (AAAI 2024 · similarity 0.85)
- A Positive-Unlabeled Metric Learning Framework for Document-Level Relation Extraction with Incomplete Labeling (AAAI 2024 · similarity 0.85)
- Enhancing Relation Extraction via Supervised Rationale Verification and Feedback (AAAI 2025 · similarity 0.85)
- Enhancing Low-Resource Relation Representations through Multi-View Decoupling (AAAI 2024 · similarity 0.84)
- Unlocking the Potential of Unlabeled Data in Semi-Supervised Domain Generalization (CVPR 2025 · 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.