MSynth: Scalable Synthetic Data Generation for Multimodal AI
MSynth develops scalable methods for generating synthetic data to train and evaluate multimodal AI systems. The project is funded by the Knut and Alice Wallenberg Foundation and runs from August 2026 to July 2028.
Umeå University is the main partner, with Lund University as collaborator. At Lund University the project is affiliated with Robotics and Semantic Systems, the LTH profile area AI and Digitalization, and ELLIIT: the Linköping-Lund initiative on IT and mobile communication. Xuan-Son Vu is the principal investigator.
Last updated: 2026-11-10