Research map / Time series / Causal and generative time series
ST-TPP: Learning Semi-Transductive Temporal Point Processes with Gromov-Wasserstein Barycentric Regularization
Qingmei Wang, Tianyu Huang, Yujie Long, Yuxin Wu, Fanmeng Wang, Xi Sun, Junchi Yan, Hongteng Xu
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 |
|---|---|---|---|
| semi-transductive prediction (add sequence and cluster embeddings) | THP · Taobao | event type accuracy: 0.477 -> 0.596 | Improved |
| semi-transductive prediction | SAHP · StackOverflow | event type accuracy: 0.021 -> 0.046 | Improved |
| semi-transductive prediction | RMTPP · Taobao | event type accuracy: 0.436 -> 0.436 | No clear effect |
| semi-transductive prediction | ODETPP · Taxi | event type accuracy: 0.866 -> 0.909 | Improved |
| semi-transductive prediction | RMTPP · Synthetic (K=2) | NMI: 0.689 -> 0.730 | Improved |
| add GWB regularizer to MLE (ablation) | THP · Synthetic (K=2) | event type accuracy: 0.296 -> 0.298 | No clear effect |
| add GWB regularizer to MLE (ablation) | THP · Synthetic (K=2) | NMI: 0.673 -> 0.689 | Improved |
| add GWB regularizer on top of MLE+CE (ablation) | THP · Synthetic (K=2) | event type accuracy: 0.306 -> 0.316 | Improved |
| add sequence-level embedding to event+cluster (ablation) | ST-THP · Synthetic (K=2) | event type accuracy: 0.293 -> 0.316 | Improved |
| add cluster-level embedding to event+sequence (ablation) | ST-THP · Synthetic (K=2) | event type accuracy: 0.309 -> 0.316 | Improved |
| increase GWB weight tau above 5e-3 (ablation) | ST-THP · Synthetic (K=2) | ELL and ACC | Hurt |
| reduce kernel sampling rate of K-tilde (ablation) | ST-THP · Synthetic (K=2) | ELL, ACC, NMI | No clear effect |
Closest papers by meaning
- Interacting Diffusion Processes for Event Sequence Forecasting (ICML 2024 · similarity 0.88)
- EasyTPP: Towards Open Benchmarking Temporal Point Processes (ICLR 2024 · similarity 0.87)
- Deep Continuous-Time State-Space Models for Marked Event Sequences (NeurIPS 2025 · similarity 0.87)
- A Plug-and-Play Bregman ADMM Module for Inferring Event Branches in Temporal Point Processes (AAAI 2025 · similarity 0.86)
- A Variational Autoencoder for Neural Temporal Point Processes with Dynamic Latent Graphs (AAAI 2024 · similarity 0.85)
- Residual TPP: A Unified Lightweight Approach for Event Stream Data Analysis (ICML 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.
