Work

Seahorse: Unified Benchmarking for Spatio-Temporal Point Processes

Year: 2026

Type: software

Abstract: A benchmark and library for spatio-temporal point processes: 13 datasets, 28 model presets, one evaluation protocol with exact likelihoods and calibration (NeurIPS 2026).

Source: Zenodo (CERN European Organization for Nuclear Research)

DOI: 10.5281/zenodo.21078077

Authors: Aalaila, Yahya, Großmann, Gerrit, Vollmer, Sebastian

Institution: German Research Centre for Artificial Intelligence

Language: English

Topic: Point processes and geometric inequalities

Subfield: Applied Mathematics

Field: Mathematics

Domain: Physical Sciences

Open Access status: green

OpenAlex ID: W7166706986

Cites:

Cited by: 1

Related to: 10