Krikri: Advancing Open Large Language Models for Greek
This addresses the need for advanced AI tools tailored to Greek, including support for Modern and Ancient Greek, but it is incremental as it builds on existing models like Llama 3.1-8B.
The researchers tackled the problem of developing a high-quality large language model for the Greek language by introducing Llama-Krikri-8B, which shows notable improvements over comparable models in natural language understanding, generation, and code generation on existing and new benchmarks.
We introduce Llama-Krikri-8B, a cutting-edge Large Language Model tailored for the Greek language, built on Meta's Llama 3.1-8B. Llama-Krikri-8B has been extensively trained on high-quality Greek data to ensure superior adaptation to linguistic nuances. With 8 billion parameters, it offers advanced capabilities while maintaining efficient computational performance. Llama-Krikri-8B supports both Modern Greek and English, and is also equipped to handle polytonic text and Ancient Greek. The chat version of Llama-Krikri-8B features a multi-stage post-training pipeline, utilizing both human and synthetic instruction and preference data, by applying techniques such as MAGPIE. In addition, for evaluation, we propose three novel public benchmarks for Greek. Our evaluation on existing as well as the proposed benchmarks shows notable improvements over comparable Greek and multilingual LLMs in both natural language understanding and generation as well as code generation.