Fundamental AI Lab

Prof. Dr. Yuki Asano

At the Fundamental AI Lab of the University of Technology Nuremberg, we are pioneering core technologies for the next generation of foundation models.

  • Our primary focus is on self-supervised learning, which allows models to learn from unstructured data, such as raw videos, without the need for manual labeling, significantly enhancing scalability. 
  • We also explore vision-language and multimodal learning, enabling AI models to seamlessly integrate visual and textual information. 
  • Furthermore, by advancing reasoning and planning in large language models (LLMs), we aim to equip AI systems with sophisticated logical thinking and decision-making abilities.

Further information and the latest news can be found on the Fundamental and AI Lab GitHub page.

Prof. Dr. Yuki Asano
Professur für Fundamental AI

Publications

Is ImageNet worth 1 video? Learning strong image encoders from 1 long unlabelled video

View Project

VeRA: Vector-based Random Matrix Adaptation

View Project

PIN: Positional Insert unlocks object localisation abilities in VLMs

View Project

SIGMA: Sinkhorn-Guided Masked Video Modeling

View Project

List of publikations of Prof. Dr. Yuki Asano on Google Scholar

List of publications of Prof. Dr. Yuki Asano on dblp

Team

Lukas Knobel

Doctoral Researcher

Dawid Kopiczko

Doctoral Researcher

Jona Ruthardt

Doctoral Researcher

Du hast Fragen?

Dann wende dich an unser Fundamental AI Lab.

 funai@utn.de