# Probabilistic Forecasting of Business Process Executions with Neural Temporal Point Processes

arXiv: https://arxiv.org/abs/2609.21382 · HTML: https://mlsanity.com/paper/2609.21382/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 27 references that are in the map.
- First posted on arXiv 18 Sep 2026.

## Builds on (most cited first, 10 of 10)

- [Transformer Hawkes Process](https://mlsanity.com/paper/2002.09291/) (2020)
- [Intensity-Free Learning of Temporal Point Processes](https://mlsanity.com/paper/1909.12127/) (2019)
- [Deep Learning for Predictive Business Process Monitoring: Review and Benchmark](https://mlsanity.com/paper/2009.13251/) (2020)
- [ProcessTransformer: Predictive Business Process Monitoring with Transformer Network](https://mlsanity.com/paper/2104.00721/) (2021)
- [A Deep Adversarial Model for Suffix and Remaining Time Prediction of Event Sequences](https://mlsanity.com/paper/2102.07298/) (2021)
- [Creating Unbiased Public Benchmark Datasets with Data Leakage Prevention for Predictive Process Monitoring](https://mlsanity.com/paper/2107.01905/) (2021)
- [Time Matters: Time-Aware LSTMs for Predictive Business Process Monitoring](https://mlsanity.com/paper/2010.00889/) (2020)
- [PGTNet: A Process Graph Transformer Network for Remaining Time Prediction of Business Process Instances](https://mlsanity.com/paper/2404.06267/) (2024)
- [What Averages Do Not Tell -- Predicting Real Life Processes with Sequential Deep Learning](https://mlsanity.com/paper/2110.10225/) (2021)
- [Directly Follows Graphs Go Predictive Process Monitoring With Graph Neural Networks](https://mlsanity.com/paper/2503.03197/) (2025)

## Abstract (arXiv)

Operators of service-based systems act on forecasts of how a running execution will continue, and such a forecast is actionable only if its reliability is known. Mainstream deep-learning models for this task are discriminative and deterministic: they emit a single next activity and a single remaining-time estimate, without a distribution to reason over. We instead cast the problem as generative sequence modelling with marked temporal point processes, which define a joint density over the next mark and its inter-event time and therefore deliver predictive distributions by construction. Real event logs violate the simple-point-process assumption these models rest on, since consecutive events frequently carry identical timestamps; we handle such ties explicitly and combine a transformer encoder with a mixture decoder over inter-event times, trained by exact log-likelihood. On ten public logs, the resulting model matches discriminative baselines on point accuracy, dominates them on the calibration and sharpness of remaining-time distributions, and is the cheapest at inference, since a full predictive distribution is obtained in a single forward pass without sampling.