PhD Research Fellow in Structured Vision-Language Models

Three-year PhD fellowship at Integreat, the Norwegian Centre for Knowledge-driven Machine Learning (a Research Council of Norway Centre of Excellence, 2023-33), in the Department of Informatics at the University of Oslo, with a possible fourth year including 25% career-promoting work. Vision-language models can name the objects in an image but often miss how they relate. This project gives such models an explicit relational representation by developing panoptic scene graphs, in which every object and background region becomes a node and edges encode spatial, semantic and relational context; the graphs are built from superpixels that follow object boundaries rather than the fixed square patches of standard vision transformers.

Starting date from January 2027 by agreement.

Master's degree or equivalent in computer science, machine learning, data science, mathematics, statistics, electrical engineering or a related field; experience in scientific programming.