Skip to main content
ProceedingsFull paper Open access

Persistent Identifiers for Model Checkpoints

In Proceedings of the International Conference on Open and Reproducible AI Research (OPENAIR 2026), pp. 13–22. Nairobi, Kenya, 19 Oct 2026.

DOI 10.xxxx/ctm.pr.2026.0012 · ID CTM-PR-2026-0012

Download PDF

Cite this record

Idrissi, F. Z. E. & Menon, R. (2026). Persistent Identifiers for Model Checkpoints. In Proceedings of the International Conference on Open and Reproducible AI Research (OPENAIR 2026) (pp. 13–22). CONTINUUM Proceedings. doi:10.xxxx/ctm.pr.2026.0012

Sample record. Fictional demonstration content.

Abstract

Model checkpoints are cited by URL and then disappear. We propose a metadata profile for model checkpoints that supports DOIs, content hashes and links to training data and code, and evaluate it by retro-fitting 120 published checkpoints.

Keywords persistent identifiers · models · metadata

Full text

CTM-PR-2026-0012.pdf

Paper · PDF · pp. 13–22

Download

Demonstration record: no file is attached.

How to cite

Idrissi, F. Z. E. & Menon, R. (2026). Persistent Identifiers for Model Checkpoints. In Proceedings of the International Conference on Open and Reproducible AI Research (OPENAIR 2026) (pp. 13–22). CONTINUUM Proceedings. doi:10.xxxx/ctm.pr.2026.0012

More from OPENAIR 2026