Research map / Graph learning / Knowledge and temporal graphs
Validation of Growing Knowledge Graphs by Abductive Text Evidences
Jianfeng Du, Jeff Z. Pan, Sylvia Wang, Kunxun Qi, Yuming Shen, Yu Deng
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 |
|---|---|---|---|
| rank by attention contribution (Formula 8) instead of normalized probability, CNN encoder (ablation) | CNNRP · DRC | average Rankmin: 25.095 -> 12.714 | Improved |
| rank by attention contribution instead of probability, CNN encoder (ablation) | CNNRP · JW | average Rankmin: 19.500 -> 4.000 | Improved |
| rank by attention contribution instead of probability, CNN encoder (ablation) | CNNRP · OM | average Rankmin: 6.173 -> 3.825 | Improved |
| rank by attention contribution instead of probability, CNN encoder (ablation) | CNNRP · RTK | average Rankmin: 8.286 -> 4.786 | Improved |
| rank by attention contribution instead of probability, PCNN encoder (ablation) | PCNNRP · DRC | average Rankmin: 20.095 -> 6.524 | Improved |
| rank by attention contribution instead of probability, PCNN encoder (ablation) | PCNNRP · JW | average Rankmin: 22.500 -> 4.500 | Improved |
| rank by attention contribution instead of probability, CNN encoder (ablation) | CNNRP · DRC | average NDCG: 0.535 -> 0.553 | Improved |
| rank by attention contribution instead of probability, CNN encoder (ablation) | CNNRP · JW | average NDCG: 0.472 -> 0.543 | Improved |
| rank by attention contribution, PCNN encoder (ablation) | PCNNRP · OM | average NDCG: 0.491 -> 0.571 | Improved |
| probability-only ranking versus random ranking | CNNP/PCNNP · JW | average Rankmin: 10.333 -> 19.500 (CNNP), 22.500 (PCNNP) | Hurt |
| probability-only ranking versus random ranking | CNNP/PCNNP · JW | average NDCG: 0.504 -> 0.472 (CNNP), 0.480 (PCNNP) | Hurt |
| use CNN vs PCNN encoder in attention ranking (ablation) | CNNRP vs PCNNRP · DRC | average Rankmin: 12.714 (CNN) vs 6.524 (PCNN) | Mixed |
Closest papers by meaning
- Heterogeneous Graph Reasoning for Fact Checking over Texts and Tables (AAAI 2024 · similarity 0.85)
- Knowledge Graph Error Detection with Contrastive Confidence Adaption (AAAI 2024 · similarity 0.84)
- Bayesian Inference with Complex Knowledge Graph Evidence (AAAI 2024 · similarity 0.84)
- Context-aware Inductive Knowledge Graph Completion with Latent Type Constraints and Subgraph Reasoning (AAAI 2025 · similarity 0.83)
- EPERM: An Evidence Path Enhanced Reasoning Model for Knowledge Graph Question and Answering (AAAI 2025 · similarity 0.83)
- NestE: Modeling Nested Relational Structures for Knowledge Graph Reasoning (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.
