Research map / Time series / Causal and generative time series
Forward-Chaining Temporal Point Process
Chao Yang, Wendi Ren, Shuang Li
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 |
|---|---|---|---|
| use true rules instead of scrambled rules (ablation) | FC-TPP · Syn@5 | Rule violation rate: 0.71 -> 0.28 | Improved |
| use true rules instead of scrambled rules (ablation) | FC-TPP · Syn@5 | KL divergence: 0.59 -> 0.42 | Improved |
| provide scrambled rules instead of no rules (ablation) | FC-TPP · Syn@5 | Rule violation rate: 0.42 -> 0.71 | Hurt |
| provide true oracle rules instead of none (ablation) | FC-TPP · Syn@5 | KL divergence: 0.54 -> 0.42 | Improved |
| add dynamic stopping for reasoning hops (ablation) | FC-TPP · MIMIC-IV | MMD: 0.48 -> 0.45 | Mixed |
| increase forward-chaining hops beyond H = 4 (ablation) | FC-TPP · Syn@5 | KL, QQ-RMSE, MMD, RV | No clear effect |
| add neural pre-train module (ablation) | FC-TPP · MIMIC-IV | DS, GPTScore, MMD: 17% in DS, 7% in GPTScore, and 4% in MMD | Improved |
| add multi-step forward chaining (ablation) | FC-TPP · MIMIC-IV | DS, GPTScore, MMD | Improved |
| add multi-round LLM rule modification and querying (ablation) | FC-TPP · MIMIC-IV | DS, GPTScore, MMD | Improved |
| use larger LLM as rule generator (TinyLlama to GPT-4o) (ablation) | FC-TPP · MIMIC-IV | MMD: 0.62 -> 0.48 | Improved |
| add multiple LLM judges to GPT-4o (ablation) | FC-TPP evaluation · MIMIC-IV | GPTScore: 0.71 -> 0.67 | No clear effect |
| reduce training data to 40% (ablation) | FC-TPP · MIMIC-IV few-shot | MMD | No clear effect |
Closest papers by meaning
- Neuro-Symbolic Temporal Point Processes (ICML 2024 · similarity 0.87)
- Latent Logic Tree Extraction for Event Sequence Explanation from LLMs (ICML 2024 · similarity 0.86)
- EasyTPP: Towards Open Benchmarking Temporal Point Processes (ICLR 2024 · similarity 0.85)
- DanmakuTPPBench: A Multi-modal Benchmark for Temporal Point Process Modeling and Understanding (NeurIPS 2025 · similarity 0.85)
- MOCHA: Discovering Multi-Order Dynamic Causal Structure in Temporal Point Processes (NeurIPS 2026 · similarity 0.85)
- Deep Continuous-Time State-Space Models for Marked Event Sequences (NeurIPS 2025 · 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.
