Research map / Label-efficient and robust learning / Noisy labels
Reinforcement Learning for Relation Classification From Noisy Data
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 reinforcement-learning instance selection before relation classification | CNN relation classifier · Sentence-level relation classification (300 manually labeled sentences) | Macro F1: 0.40 -> 0.42 | Helped |
| Add reinforcement-learning instance selection before relation classification | CNN relation classifier · Sentence-level relation classification (300 manually labeled sentences) | Accuracy: 0.60 -> 0.64 | Helped |
| Use reinforcement-learning instance selection instead of bag-level max selection | CNN+Max · Sentence-level relation classification (300 manually labeled sentences) | Macro F1: 0.06 -> 0.42 | Helped |
| Use reinforcement-learning instance selection instead of bag-level max selection | CNN+Max · Sentence-level relation classification (300 manually labeled sentences) | Accuracy: 0.34 -> 0.64 | Helped |
| Use reinforcement-learning instance selection instead of bag-level attention | CNN+ATT · Sentence-level relation classification (300 manually labeled sentences) | Macro F1: 0.29 -> 0.42 | Helped |
| Use reinforcement-learning instance selection instead of bag-level attention | CNN+ATT · Sentence-level relation classification (300 manually labeled sentences) | Accuracy: 0.56 -> 0.64 | Helped |
| Replace reinforcement-learning selection with greedy top-N selection by a pretrained CNN (ablation) | CNN+RL instance selector · Held-out relation classification evaluation | Relation classification performance | Helped |
| Train the CNN relation classifier on selector-chosen sentences rather than the original data | CNN and CNN+ATT · Held-out relation classification evaluation | Relation classification performance | Helped |
Closest papers by meaning
- Large Scaled Relation Extraction With Reinforcement Learning (AAAI 2018 · similarity 0.91)
- A Hierarchical Framework for Relation Extraction with Reinforcement Learning (AAAI 2019 · similarity 0.90)
- Improving Neural Relation Extraction with Positive and Unlabeled Learning (AAAI 2020 · similarity 0.89)
- Joint Entity and Relation Extraction with a Hybrid Transformer and Reinforcement Learning Based Model (AAAI 2020 · similarity 0.88)
- Enhancing Relation Extraction via Supervised Rationale Verification and Feedback (AAAI 2025 · similarity 0.83)
- Relation-R1: Progressively Cognitive Chain-of-Thought Guided Reinforcement Learning for Unified Relation Comprehension (AAAI 2026 · similarity 0.83)
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.