DNA 1.0 Technical Report (2501.10648v1)

Abstract: In this report, we present DNA 1.0 8B Instruct, a state-of-the-art bilingual LLM optimized for Korean and English language tasks. By applying continual pre-training (CPT) with high-quality Korean datasets to Llama 3.1 8B and subsequent supervised fine-tuning (SFT), we create an instruction-following model with enhanced Korean language capabilities. This model is then merged with Llama 3.1 8B Instruct via spherical linear interpolation (SLERP) and undergoes further optimization through direct preference optimization (DPO) and knowledge distillation (KD). DNA 1.0 8B Instruct achieves state-of-the-art results on Korean-specific tasks, including KMMLU (53.26%), KoBEST (83.40%), and BELEBELE (57.99%), while maintaining strong English capabilities on MMLU (66.64%), MMLU-Pro (43.05%) and GSM8K (80.52%). As an open model, DNA 1.0 8B Instruct represents a significant advancement in bilingual language modeling. As an open model, DNA 1.0 8B Instruct is freely available through https://huggingface.co/dnotitia/Llama-DNA-1.0-8B-Instruct . For commercial licensing inquiries or feedback, please contact us at https://www.dnotitia.com/contact/post-form