Research map / Time series / Causal and generative time series
ITPP: Learning Disentangled Event Dynamics in Marked Temporal Point Processes
Wang-Tao Zhou, Zhao Kang, Ke Yan, Ling Tian
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 channel-independent encoding/decoding instead of channel mixing (ablation) | ITPP · StackOverflow and Earthquake | likelihood and prediction metrics | Improved |
| add type-aware inverted self-attention across channels (ablation) | ITPP · StackOverflow and Earthquake | likelihood and prediction metrics | Improved |
| use ITPP over best baseline in joint NLL | ODE-based MTPP · StackOverflow | TM-NLL: 2.198 -> 2.106 | Improved |
| use ITPP over best baseline in joint NLL | ODE-based MTPP · Taobao | TM-NLL: -1.176 -> -1.189 | Improved |
| use ITPP over NHP for mark prediction | ODE-based MTPP · MIMIC | F1: 0.850 (NHP) vs 0.856 (ITPP) | Improved |
| use self-attention encoders (THP, SAHP, AttNHP) instead of RNNs | MTPP baselines · StackOverflow | TM-NLL: THP 2.325 vs NHP 2.233 | No clear effect |
| use channel-mixing ODE models (NeuralODE, ODE-GRU) without channel independence | ODE-based MTPP · Earthquake | overfitting | Hurt |
| add channel independence and inverted attention for overfitting resistance | ITPP · Earthquake | test loss | Improved |
Closest papers by meaning
- Residual TPP: A Unified Lightweight Approach for Event Stream Data Analysis (ICML 2025 · similarity 0.86)
- Decoupled Marked Temporal Point Process using Neural Ordinary Differential Equations (ICLR 2024 · similarity 0.86)
- Deep Continuous-Time State-Space Models for Marked Event Sequences (NeurIPS 2025 · similarity 0.86)
- A Plug-and-Play Bregman ADMM Module for Inferring Event Branches in Temporal Point Processes (AAAI 2025 · similarity 0.85)
- Arbitrarily Conditioned Hierarchical Flows for Spatiotemporal Events (NeurIPS 2026 · similarity 0.85)
- Hyper Hawkes Processes: Interpretable Models of Marked Temporal Point Processes (NeurIPS 2026 · 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.
