PhD Research Fellow in Structured Vision-Language Models

The University of Oslo's Integreat centre is seeking a motivated PhD research fellow to work on structured vision-language models within an interdisciplinary machine learning environment. The project focuses on building panoptic scene graphs using superpixel tokenization, fusing visual relational data with large language models to enhance reasoning and reduce hallucination. The successful candidate will conduct independent research, collaborate with senior researchers across partner institutions, and contribute to teaching or research assistance if a fourth-year extension is undertaken. The position requires strong academic credentials, advanced programming skills, and a commitment to rigorous scientific publication.

• Develop superpixel-based panoptic segmentation models built on modern transformer architectures.
• Design methods that predict relations between segmented regions and generate panoptic scene graphs.
• Fuse scene graphs with large language models for visual question answering and grounded reasoning, including graph visual RAG.
• Publish and present research findings at leading machine learning and computer vision venues.
• Participate in collaborative research activities across the interdisciplinary centre.