Researchers pretrain compact Greek-only language models from scratch
Two decoder-only transformers, LexisLM-1B (1.09B parameters) and LexisLM-500M (502M), were trained from scratch exclusively on Greek text, per a new arXiv paper.
Read the full story at arXiv cs.CL ↗
Timeline · 1 report
- 2026-11-10 05:00 · arXiv cs.CL
Compact Greek Language Models from Scratch: Pretraining 500M- and 1B-Parameter Decoder-Only Transformers