Metraipolis › archive › dc_ad608e0362ff

Compact Greek Language Models from Scratch: Pretraining 500M- and 1B-Parameter Decoder-Only Transformers — arXiv:2611.10209

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.10209
License (as archived)
CC0-1.0
Archived
2026-11-10T18:44:24.131Z
SHA-256 of the archived text
bb702445472b8c4913c82727fe985c1b86be6342cae410a1cf35cc68cc9063f2
Stable URL of this page
https://metraipolis.com/documento/dc_ad608e0362ff

How to cite

Compact Greek Language Models from Scratch: Pretraining 500M- and 1B-Parameter Decoder-Only Transformers — arXiv:2611.10209 — arXiv (en), https://arxiv.org/abs/2611.10209 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ad608e0362ff (Metraipolis archive dc_ad608e0362ff, sha256 bb702445472b8c4913c82727fe985c1b86be6342cae410a1cf35cc68cc9063f2)
BibTeX
@misc{dcad608e0362ff,
  title = {Compact Greek Language Models from Scratch: Pretraining 500M- and 1B-Parameter Decoder-Only Transformers — arXiv:2611.10209},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.10209},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ad608e0362ff; sha256 bb702445472b8c4913c82727fe985c1b86be6342cae410a1cf35cc68cc9063f2},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.10209. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We present two compact decoder-only language models trained from scratch exclusively on Greek text: LexisLM-1B (1.09B parameters) and LexisLM-500M (502M parameters). Both use a dedicated 32k byte-level BPE tokenizer and a 22B-token pretraining corpus drawn primarily from Greek Wikipedia and FineWeb2. Both models complete the 22B-token schedule, with best held-out perplexities of 9.77 for LexisLM-1B and 11.31 for LexisLM-500M. Evaluation of both models on Greek-translated ARC, MMLU, and TruthfulQA shows above-chance performance on ARC-Easy, while ARC-Challenge and MMLU remain near chance. We also perform supervised fine-tuning of both models on Greek instruction data, primarily EuroAlpaca, reaching held-out instruction perplexities of 7.12 for LexisLM-500M and 6.37 for LexisLM-1B. The report describes the tokenizer, data preparation, training, and evaluation methods, providing a foundation for further work on compact Greek language models.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2611.10209 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-10T18:37:52.420Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).