Understanding causality in event sequences with thousands of sparse event types is critical in domains such as healthcare, cybersecurity, or vehicle diagnostics, yet current methods fail to scale. We present OSCAR, a one-shot causal autoregressive method that infers per-sequence Markov Boundaries using two pretrained Transformers as density estimators. This enables efficient, parallel causal discovery without costly global CI testing. On a real-world automotive dataset with 29,100 events and 474 labels, OSCAR recovers interpretable causal structures in minutes, while classical methods fail to scale, enabling practical scientific diagnostics at production scale.
@article{arxiv.2509.23213,
title = {One-Shot Multi-Label Causal Discovery in High-Dimensional Event Sequences},
author = {Hugo Math and Robin Schön and Rainer Lienhart},
journal= {arXiv preprint arXiv:2509.23213},
year = {2025}
}
Comments
Accepted at NeuRIPS2025 Workshop CauScien: Discovering Causality in Science. arXiv admin note: substantial text overlap with arXiv:2509.19112