Tech Blog

Getting AI to understand the field: NAVER Cloud’s forward deployed engineers
Ask an AI to find a work document, and it answers. Ask it to analyze some data, and it produces results. So is putting AI to work on real tasks just a matter of having a model ready? On the ground, the story is a little different. The material you need is scattered across several […]
Oct 26, 2026
• # FDE
• # Harness
• # NAVERCloud
• # Ontology
• # RAG

NAVER Cloud Defense AI Day: Insights on sovereign AI strategy for national defense
Oct 6, 2026
• # Defense AI
• # NAVERCloud
• # sovereignAI

Training an image generation model from scratch: What we chose, and how it fits together
Sep 29, 2026
• # Image editing
• # Image generation

Unlocking the potential of the latest GPUs: Optimizing Mamba-2 kernels for the B200
Sep 4, 2026
• # kernel optimization
• # Mamba-2

Speed and cost: The “image token imbalance” eating into multimodal AI training
Aug 13, 2026
• # HyperCLOVA X

Conversation doesn’t wait its turn: Sommelier, a data pipeline for real-time conversational voice AI
Jul 16, 2026
• # ACL
• # AI Research
• # full-duplex model
• # NAVER Cloud
• # open source
• # Sommelier

Split the conflict, merge the gains: MERIT, an efficient post-training strategy for LLMs
Jul 1, 2026
• # AI Research
• # Gradient Conflict
• # NAVER Cloud

Seeing and hearing the field for ourselves: Building our own encoders and HyperCLOVA X SEED 4B
Jun 23, 2026
• # AudioLLM
• # foundation model
• # Fromscratch
• # HyperCLVOA X
• # NAVER Cloud
• # NAVER LLM
• # Omni-modality
• # Omnimodality
• # Vision encoder

Hallucinations leave a trace: Detecting LLM hallucination with future context
Jun 11, 2026
• # AI Safety

NAVER Cloud joins the NVIDIA Nemotron Coalition: Collaborating toward sovereign AI
Jun 10, 2026
• # GTC
• # HyperCLVOA X
• # NAVER Cloud
• # Nemotron
• # Nemotron Coalition
• # NVIDIA
• # sovereignAI