{"ID":2863591,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.24762","arxiv_id":"2509.24762","title":"In-Context Learning of Temporal Point Processes with Foundation Inference Models","abstract":"Modeling event sequences of multiple event types with marked temporal point processes (MTPPs) provides a principled way to uncover governing dynamical rules and predict future events. Current neural network approaches to MTPP inference rely on training separate, specialized models for each target system. We pursue a radically different approach: drawing on amortized inference and in-context learning, we pretrain a deep neural network to infer, in-context, the conditional intensity functions of event histories from a context defined by sets of event sequences. Pretraining is performed on a large synthetic dataset of MTPPs sampled from a broad distribution of Hawkes processes. Once pretrained, our Foundation Inference Model for Point Processes (FIM-PP) can estimate MTPPs from real-world data without any additional training, or be rapidly finetuned to target systems. Experiments show that this amortized approach matches the performance of specialized models on next-event prediction across common benchmark datasets.","short_abstract":"Modeling event sequences of multiple event types with marked temporal point processes (MTPPs) provides a principled way to uncover governing dynamical rules and predict future events. Current neural network approaches to MTPP inference rely on training separate, specialized models for each target system. We pursue a ra...","url_abs":"https://arxiv.org/abs/2509.24762","url_pdf":"https://arxiv.org/pdf/2509.24762v2","authors":"[\"David Berghaus\",\"Patrick Seifner\",\"Kostadin Cvejoski\",\"César Ojeda\",\"Ramsés J. Sánchez\"]","published":"2025-09-29T13:28:06Z","proceeding":"cs.LG","tasks":"[\"cs.LG\"]","methods":"[]","has_code":false}