Research map / Learning theory and algorithms / Combinatorial and search optimization
Numerical Relation Extraction with Minimal Supervision
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 keyword-based labeling to KB distant supervision (ablation) | NumberTron (Atleast-K) · Numerical relation extraction | F1: 45.97 -> 63.78 | Helped |
| add KB supervision to keyword-only labeling (ablation) | NumberTron (Atleast-K) · Numerical relation extraction | F1: 54.83 -> 63.78 | Helped |
| use Atleast-K instead of Simple OR join potential (ablation) | NumberTron · Numerical relation extraction | F1: 63.21 -> 63.78 | Helped |
| use Agreeing-K instead of Atleast-K join potential (ablation) | NumberTron · Numerical relation extraction | F1: 63.78 -> 61.79 | Hurt |
| add keyword features to Mintz features (ablation) | NumberTron · Numerical relation extraction | F1: 28.21 -> 42.71 | Helped |
| add number features to Keyword+Mintz (ablation) | NumberTron · Numerical relation extraction | F1: 42.71 -> 63.78 | Helped |
| add Mintz features to Keyword+Number (ablation) | NumberTron · Numerical relation extraction | F1: 54.80 -> 63.78 | Helped |
| replace rule-based NumberRule with statistical NumberTron (ablation) | extractor · Numerical relation extraction | recall: 53.60 -> 66.92 | Helped |
| replace NumberRule with NumberTron | extractor · Numerical relation extraction | F1: 56.30 -> 63.78 | Helped |
| NumberTron vs NumberRule per relation | extractor · Numerical relation extraction | F1 per relation: FDI 0 vs 50.00; Inflation 88.40 vs 56.25 | Mixed |
| NumberTron vs MultiR++ | extractor · Numerical relation extraction | F1: 38.84 -> 63.78 | Helped |
Closest papers by meaning
- Global Distant Supervision for Relation Extraction (AAAI 2016 · similarity 0.86)
- COMM: Concentrated Margin Maximization for Robust Document-Level Relation Extraction (AAAI 2025 · similarity 0.84)
- Enhancing Relation Extraction via Supervised Rationale Verification and Feedback (AAAI 2025 · similarity 0.84)
- A Hierarchical Framework for Relation Extraction with Reinforcement Learning (AAAI 2019 · similarity 0.84)
- Prototypical Representation Learning for Relation Extraction (ICLR 2021 · similarity 0.83)
- Reliable Data Generation and Selection for Low-Resource Relation Extraction (AAAI 2024 · 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.