Speaker
Description
Since 2023, Japan has significantly intensified its research and development activities concerning foundation models, particularly large language models (LLMs). These efforts have encompassed not only independent initiatives within the private sector but also proactive governmental support aimed at fostering collaboration among industry and academia. LLM-jp is an academia-focused research consortium led primarily by the National Institute of Informatics (NII), undertaking comprehensive research and development on large language models and multimodal models with government backing. Notably, the consortium facilitates projects that are challenging for individual laboratories to pursue independently, including the training of models at the 100-billion-parameter scale, which require extensive computational resources. It also serves as a forum for disseminating the outcomes of these initiatives. In this keynote session, we will present an overview of these recent advancements.
