LLMC Symposium
2nd Japanese Symposium on Open Large Language Models
Challenges for Open LLM Development
Networking Session 18:00–20:00
R&D Hub Aimed at Ensuring Transparency and Reliability of Generative AI Models
Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan
OVERVIEW
About the Symposium
The rapid advancement of generative AI is transforming research, industry, and society. As large language models (LLMs) continue to grow in scale and capability, ensuring their transparency, reliability, and safety and promoting responsible AI development and open research have become an important global challenge.
This symposium will focus on the challenges surrounding the research and development of open LLMs. Researchers, developers, participants from companies and other organizations, and members of the open-source community in Japan and abroad will share their latest research findings and insights.
Through keynote lectures by international researchers, presentations on the activities and achievements of LLMC and LLM-jp, poster presentations, and a fireside chat with the keynote speakers, we will explore the current state and future directions of open LLM development.
PROGRAM
Program
Opening Remarks
President, Research Organization of Information and Systems (ROIS) / Ministry of Education, Culture, Sports, Science and Technology (TBC)
Overview of LLMC and LLM-jp Activities
KUROHASHI Sadao →
Director-General, National Institute of Informatics;
Director, Research and Development Center for Large Language Models
LLM-jp Research Highlights
Presentations by the LLM-jp working groups on their latest research activities and achievements.
-
Corpora Construction WG
KAWAHARA Daisuke →
Professor, Waseda University
Corpus Construction for Open Japanese Large Language Models -
Model Construction WG
SUZUKI Jun →
Professor, Tohoku University
Large Language Model Development at LLMC -
Tuning & Evaluation WG
MIYAO Yusuke →
Professor, The University of Tokyo
Post-training and Evaluation of LLM-jp Models -
Multimodal WG
OKAZAKI Naoaki →
Professor, Institute of Science Tokyo
Building the Vision-Language Model LLM-jp-4-VL -
Real-world Environment Interaction WG
OGATA Tetsuya →
Professor, Waseda University
Advancing Robot Foundation Models: Video-Based Trajectory Extraction and Data-Efficient Imitation Learning -
Dialogues WG
HIGASHINAKA Ryuichiro →
Professor, Nagoya University
Development of Japanese Spoken Dialogue Models and Collection of Dialogue Data -
Scientific Domain WG
AIZAWA Akiko →
Professor, National Institute of Informatics
Title to be announced -
Safety WG
SEKINE Satoshi →
Project Professor, National Institute of Informatics
AI Safety Research at NII/LLMC -
Principles Elucidation WG
OSEKI Yohei →
Associate Professor, The University of Tokyo
Interpreting and Steering Large Language Models
Lunch Break
EuroLLM, Dynamic Sparsity, and Beyond: Open Multilingual LLMs for Europe and Challenges Ahead
André Martins →
Instituto Superior Técnico, Portugal
Break
Break & Poster Session
Title to be announced
Speaker to be announced
Swiss AI Initiative
Fireside Chat with the Keynote Speakers
Panelists: André Martins / Niyu Ge / Wulong Liu / Speaker to be announced
Moderator: Pontus Stenetorp
Project Professor, National Institute of Informatics
Closing Remarks
Poster Session
Networking Session
The networking session requires a separate fee and advance registration.