PhD position Compositionality Computer Vision
Listed on 2026-07-07
-
IT/Tech
AI Business & Operations, Computer Science, Data Scientist, Machine Learning/ ML Engineer
PhD position on Compositionality for Computer Vision
The Data Management and Biometrics (DMB) group at the University of Twente is seeking one PhD candidate to work on compositional learning for data‑efficient vision foundation models under the supervision of Dr. Nicola Strisciuglio.
About the projectFoundation models in computer vision rely on massive datasets and brute‑force scaling, leading to high data requirements and hidden biases. This project aims to develop data‑efficient and reliable training strategies that reduce the need for large datasets and improve robustness.
The project will explore strategies related to:
Object‑attribute compositionality to replace exhaustive data requirements with structured concept learning; bias detection and machine unlearning to identify and mitigate bias early during training; perceptual and conceptual priors to design self‑supervised objectives that capture continuous similarity rather than binary contrastive notions.
By embedding compositional structure and prior knowledge into the training process, the project aims to break the dependency on uncontrolled large‑scale datasets and enable broader, more transparent development of vision foundation models.
About the PhD positionWe offer a fully funded 4‑year PhD position focused on compositional learning strategies for vision foundation models. The aim is to explore compositional learning strategies (e.g., attribute‑object visual representation or other forms) to reduce the data requirements needed to train vision foundation models.
When applying, you are required to clearly motivate why you are a good fit for this position and how your expertise and previous experience match the position topic.
Your profile- You have (or will shortly acquire) a master’s degree in Computer Science, Computer Engineering, or Mathematics with a major in Artificial Intelligence, Machine Learning, or Computer Vision.
- Background in machine learning, particularly in computer vision, solid understanding of mathematics and excellent programming skills.
- Interest in research and developing creative solutions to advance compositional learning and data efficiency for foundation models for vision.
- Creative mindset and excellent analytical and communication skills.
- Good team spirit and willingness to work in an interdisciplinary and internationally oriented environment.
- Proficiency in English.
- Previous experience in scientific publishing (at top venues like CVPR, ICCV, ECCV, ICLR, etc.) is appreciated.
- Full‑time position for four years, with a qualifier in the first year and flexibility to work partially from home.
- Salary and conditions in accordance with the collective labour agreement for Dutch universities (CAO‑NU).
- Gross monthly salary ranging from €3,059 (first year) to €3,881 (fourth year).
- Benefits include a holiday allowance of 8% of gross annual salary, an end‑of‑year bonus of 8.3%, and a solid pension scheme.
- A minimum of 232 leave hours for full‑time employment based on a 38‑hour week (extrapolated to 96 extra leave hours for a 40‑hour week in practice).
- Free access to sports facilities on campus.
- Family‑friendly institution offering parental leave (paid and unpaid).
- Training programme as part of the Twente Graduate School, including a plan for suitable education and supervision, and a travel budget to attend conferences and scientific events.
- Submit your application before July 21, 2026: CV (max 2 pages A4), cover letter (max 1 page A4) emphasising qualifications and motivations, academic transcript, and an IELTS/TOEFL/CAE‑C test if required for non‑English qualification holders.
First round interviews will be held during the second half of August 2026. Screening is part of the selection process.
For more information, contact Dr. Nicola Strisciuglio (nicola.strisciuglio).
About the organisationThe faculty of Electrical Engineering, Mathematics and Computer Science (EEMCS) uses mathematics, electronics and computer technology to contribute to the development of information and communication technology. The faculty collaborates with industrial partners and researchers in the Netherlands and abroad, and engages in extensive research for external commissioners and funders.
#J-18808-LjbffrTo Search, View & Apply for jobs on this site that accept applications from your location or country, tap here to make a Search: