LLM research and development initiatives at the National Institute of Informatics @WeightsBiases
LLM research and development initiatives at the National Institute of Informatics  @WeightsBiases
Uploaded December 2025 | Updated September 2026, 2 weeks ago
The National Institute of Informatics established the Large Language Model Research and Development Center in April last year. We are leading the llm-jp (LLM Study Group) activities, which have been ongoing since 2023, aiming to build LLMs with high performance on Japanese tasks and enhanced transparency and reliability. This presentation will introduce specific examples, focusing on models we have released so far, including corpus construction for training, training from scratch, tuning and evaluation, and multimodal capabilities.

--

国立情報学研究所では昨年4月に大規模言語モデル研究開発センターを開設し、2023年より継続しているllm-jp(LLM勉強会)の活動を主宰する形で、日本語タスク性能が高く、透明性・信頼性を向上させたLLMの構築を目指して研究開発に取り組んでいます。本講演では、これまでに公開したモデルを中心に、学習コーパス構築、スクラッチからの学習、チューニング・評価、マルチモーダル対応などの具体例を紹介します。
LLM research and development initiatives at the National Institute of InformaticsWhat Makes LLM Inference So HardA T cell foundation model for AI-powered target discovery and precision medicineIntroducing serverless reinforcement learning: Train reliable AI agents without worrying about GPUsBuild reliable AI agents using W&B TrainingThe FDA is slowing American companies?The Hidden Problem in Scaling LLM InferenceReal-time speech translation is changing everythingYou Can Log Videos to Your W&B RunsHow Runway Reached #1 in Video AIWhere AI Outworks HumansSynthetic data in medical device AI: Challenges and opportunities
Weights & Biases |

LLM research and development initiatives at the National Institute of Informatics

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER