Research map / Probabilistic and scientific modeling / Causal discovery
Beyond Entity Correlations: Disentangling Event Causal Puzzles in Temporal Knowledge Graphs
Qian Chen, Jinyu Zhang, Ling Chen
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 counterfactual detector module with contrastive loss (ablation) | HEDRA · ICEWS14 | MRR: 47.11 -> 47.86 | Improved |
| add event importance to non-causality mask (ablation) | HEDRA · ICEWS14 | MRR: 47.34 -> 47.86 | Improved |
| add distributional discrepancy to non-causality mask (ablation) | HEDRA · ICEWS14 | MRR: 47.25 -> 47.86 | Improved |
| add IV-guided disentangling module with robustness loss (ablation) | HEDRA · ICEWS14 | MRR: 46.47 -> 47.86 | Improved |
| add evolutionary orthogonal module with evolutionary loss (ablation) | HEDRA · ICEWS14 | MRR: 46.24 -> 47.86 | Improved |
| add IV-guided disentangling module with robustness loss (ablation) | HEDRA · ICEWS14 | Hits@1: 33.77 -> 35.28 | Improved |
| add evolutionary orthogonal module with evolutionary loss (ablation) | HEDRA · ICEWS14 | Hits@10: 74.10 -> 75.65 | Improved |
| use HEDRA instead of DECRL | HEDRA · ICEWS14 | MRR: 42.90 -> 47.86 | Improved |
| use HEDRA instead of RE-NET | HEDRA · ICEWS18 | Hits@10: 75.66 -> 75.64 | No clear effect |
| use HEDRA instead of DECRL (training cost) | HEDRA · ICEWS14 | training time (s): 907.47 -> 1226.63 | Hurt |
| increase number of neighbors k in candidate graph (ablation) | HEDRA · ICEWS14 | MRR | Mixed |
| vary history window length (ablation) | HEDRA · ICEWS14 | MRR | No clear effect |
Closest papers by meaning
- Disentangled Representation Learning in Non-Markovian Causal Systems (NeurIPS 2024 · similarity 0.87)
- MOCHA: Discovering Multi-Order Dynamic Causal Structure in Temporal Point Processes (NeurIPS 2026 · similarity 0.87)
- Local Identifying Causal Relations in the Presence of Latent Variables (ICML 2025 · similarity 0.86)
- CausalTab: Pretraining Across Causal Environments for Tabular Causal Discovery (NeurIPS 2026 · similarity 0.86)
- TEILP: Time Prediction over Knowledge Graphs via Logical Reasoning (AAAI 2024 · similarity 0.85)
- Causal Representation Learning via Counterfactual Intervention (AAAI 2024 · similarity 0.85)
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.
