JABERT

Citation and License

Full details regarding the training and comparative evaluation of JABERT can be found in our paper in LT4HALA 2026 (see citation below).

This model is released under the open Creative Commons Attribution 4.0 International (CC BY 4.0) license, allowing both academic and commercial use.

Please cite

For all use or reuse of this model, please cite:

Elisha Rosensweig, Yitzchak Lindenbaum, Hillel Gershuni, Vered Raziel-Kretzmer, Daniel Caine, Avi Shmidman, "JABERT: A New State-of-the-Art BERT Model for Judeo-Arabic",  
Proceedings of the 2026 Workshop on Language Technologies for Historical and Ancient Languages (LT4HALA), forthcoming.

Bibtex:

@inproceedings{rosensweig-etal-2026-jabert,
  author    = {Elisha Rosensweig and Yitzchak Lindenbaum and Hillel Gershuni and Vered Raziel-Kretzmer and Daniel Caine and Avi Shmidman},
  title     = {JABERT: A New State-of-the-Art BERT Model for Judeo-Arabic},
  booktitle = {Proceedings of the 2026 Workshop on Language Technologies for Historical and Ancient Languages (LT4HALA)},
  year      = {2026},
  note      = {Forthcoming}
}

Funding

This work has been funded by the European Union (ERC, MiDRASH, Project No. 101071829; principal investigators: Avi Shmidman, Bar-Ilan University; Nachum Dershowitz, Tel Aviv University; Judith Olszowy-Schlanger, EPHE-PSL; and Daniel Stökl Ben Ezra, EPHE-PSL), for which we are grateful.

Views and opinions expressed are, however, those of the authors only and do not necessarily reflect those of the European Union or the European Research Council Executive Agency. Neither the European Union nor the granting authority can be held responsible for them.