Challenge Overview: The primary goal of this challenge is to advance the capability of Large Vision-Language Models to accurately interpret and understand complex visual data such as Data Flow Diagrams (DFDs), Class Diagrams, Gantt Charts, and Building Design Drawings. We invite AI researchers, data scientists, and practitioners with interest and experience in natural language processing, computer vision, and multimodal learning to join this grand challenge. Participants can register as individuals or teams. They are required to develop a model that can answer questions related to the input data. This year, we invite participants to solve the problem of Japanese government-related and business documents. Each document will be in pdf format which is associated with 10-choice question.
Register your team here!.

Datasets:

• Public dataset: We re-annontate publicly available JDocQA dataset,resulting in about 300 samples. The annotation format is provided later.
• Private dataset: We collect data by ourselves and annotate in the same manner with the public data.

We note that all data are for research purposes only under CC-BY-NC (商用利用不可) license. We appreciate your understanding and cooporation.

Prizes and Travel Grants:Travel grants and Prizes will be announced later.
Computational Resources:Participants from the University of Tokyo may use SoftBank Beyond AI SANDBOX GPUs.