Latvian DeBERTaV3 base model

Latvian DeBERTaV3 text encoder model trained with a replaced token detection (RTD) objective, released with the paper "Pretraining and Benchmarking Modern Encoders for Latvian".

For evaluation code and benchmark results, see: https://github.com/LUMII-AILab/latvian-encoders

Citation

@inproceedings{znotins-2026-modern_lv_encoders,
    title = "Pretraining and Benchmarking Modern Encoders for {L}atvian",
    author = "Znotins, Arturs",
    booktitle = "Proceedings of the Second Workshop on Language Models for Low-Resource Languages ({LoResLM})",
    year = "2026",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2026.loreslm-1.40/",
    pages = "461--470"
}

Acknowledgements

This work was supported by the EU Recovery and Resilience Facility project Language Technology Initiative (2.3.1.1.i.0/1/22/I/CFLA/002).

AiLab-IMCS-UL/lv-deberta-base

Latvian DeBERTaV3 base model

Citation

See also

Acknowledgements