Research map / Large language models / Retrieval and knowledge
Integrating Deep Learning with Logic Fusion for Information Extraction
Wenya Wang, Sinno Jialin Pan
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 |
|---|---|---|---|
| Add probabilistic logic fusion to the transformer | TransF · Restaurant14, Laptop14, Restaurant16 (OTE) | F1: 84.64 -> 85.62; 81.76 -> 82.46; 73.56 -> 74.67 | Improved |
| Add logic fusion to the transformer | TransF · TREC without boundary | Entity F1: 85.8 -> 87.1 | Improved |
| Add logic fusion to the transformer | TransF · TREC without boundary | Relation F1: 62.7 -> 64.6 | Improved |
| Add segmentation rules (ablation) | TransF · Restaurant14, Laptop14, Restaurant16 (OTE) | F1: 84.6 -> 85.0; 81.8 -> 82.1; 73.6 -> 73.9 | Improved |
| Add segmentation rules (ablation) | TransF · ACE05 | Entity F1: 83.2 -> 83.2 | No clear effect |
| Add segmentation and entity-relation rules (ablation) | TransF · TREC | Relation F1: 62.7 -> 64.6 | Improved |
| Add segmentation and entity-relation rules (ablation) | TransF · ACE05 | Relation F1: 59.1 -> 59.3 | No clear effect |
| Add global-loss fine-tuning (ablation) | Ours · ACE05 | Relation F1: 59.3 -> 59.4 | No clear effect |
| Remove attention-weight features (ablation) | TransF · TREC | Relation F1: 62.7 -> 60.6 | Hurt |
| Remove attention-weight features (ablation) | TransF · ACE05 | Relation F1: 59.1 -> 57.4 | Hurt |
| Add label embeddings (ablation) | TransF without label embeddings · Restaurant14, Laptop14, Restaurant16 (OTE) | F1: 84.1 -> 84.6; 77.6 -> 81.8; 72.7 -> 73.6 | Improved |
| Remove POS embeddings (ablation) | TransF · TREC | Entity F1: 87.1 -> 85.0 | Hurt |
Closest papers by meaning
- Joint Entity and Relation Extraction with a Hybrid Transformer and Reinforcement Learning Based Model (AAAI 2020 · similarity 0.83)
- Differentiable Learning of Logical Rules for Knowledge Base Reasoning (NeurIPS 2017 · similarity 0.82)
- Enhancing Multi-Hop Fact Verification with Structured Knowledge-Augmented Large Language Models (AAAI 2025 · similarity 0.81)
- Rethinking Complex Queries on Knowledge Graphs with Neural Link Predictors (ICLR 2024 · similarity 0.81)
- RuAG: Learned-rule-augmented Generation for Large Language Models (ICLR 2025 · similarity 0.81)
- A Reasoning Paradigm for Named Entity Recognition (AAAI 2026 · similarity 0.81)
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.
