About Me

I am an ELLIS PhD student, co-supervised by Prof. Dr. Bernt Schiele from the Max Planck Institute for Informatics and Dr. Federico Tombari from Google Zurich.

Previously, I was a Student Researcher at Google Zurich between September 2025 and May 2026, working in Dr. Federico Tombari’s team under the supervision of Dr. Ferjad Naeem on elastic inference with LVLMs.

I obtained my BS in Computer Engineering from Middle East Technical University where I was advised by Prof. Dr. Sinan Kalkan. I was also lucky enough to be an intern at Five AI (Bosch) Oxford Applied Research Center under the supervision of Dr. Kemal Oksuz and Dr. Puneet K. Dokania, at the University of Cambridge, Affective Intelligence and Robotics Lab under the supervision of Prof. Hatice Gunes and also at the METU Image Lab under the supervision of Prof. Dr. Sinan Kalkan and Prof. Emre Akbas.

Previously, I worked broadly on robustness, reliability and fairness of machine learning models. Nowadays, I am interested in foundation models, with a special focus on efficient and elastic vision-language modeling for building generalist models capable of operating under different resource constraints.

Experience

Max Planck Institute for Informatics

Doctoral Researcher

Efficient foundation models, Vision-Language Modeling

Saarbruecken, Germany, Sep 2024 - Present

Google Zurich

Student Researcher

Efficient and Elastic Large Vision-Language Models

Zurich, Switzerland, Sep 2025 - May 2026

Five AI (Bosch)

Intern Research Scientist

Robustness and Calibration of Object Detectors

Oxford, The UK, Sep 2023 - Jun 2024

University of Cambridge, Affective Intelligence and Robotics Lab

Undergraduate Researcher

Fairness Through Uncertainty, Facial Activation Unit Recognition Through GNNs

Cambridge, The UK, Feb 2022 - Aug 2023

METU Image Lab

Undergraduate Researcher

Quantifying and Mitigating Class Imbalance Through Uncertainty

Ankara, Turkey, Oct 2021 - Aug 2023

Selected Publications

• Matteo Farina*, Vishaal Udandarao*, Thao Nguyen*, Selim Kuzucu, et al., DataComp-VLM: Improved open datasets for Vision-Language Models. preprint. arXiv, Project Page.

• Selim Kuzucu, Alessio Tonioni, Vasile Lup, Bernt Schiele, Federico Tombari, M. Ferjad Naeem. PARCEL: Pool-Anchored Resampling with Conditioned Elastic Queries for Efficient Vision-Language Understanding. preprint. arXiv, Project Page.

• Selim Kuzucu, M. Ferjad Naeem, Anna Kukleva, Federico Tombari, Bernt Schiele. Language-Unlocked ViT (LUViT): Empowering Self-Supervised Vision Transformers with LLMs. preprint. arXiv.

• Selim Kuzucu, Kemal Oksuz, Jonathan Sadeghi, Puneet K. Dokania. On Calibration of Object Detectors: Pitfalls, Evaluation and Baselines. Published at the European Conference on Computer Vision (ECCV 2024) as an oral presentation (top 2.3% of valid submissions). arXiv.

• Kemal Öksüz, Selim Kuzucu, Tom Joy, Puneet K. Dokania. MoCaE: Mixture of Calibrated Experts Significantly Improves Object Detection. Published at Transactions on Machine Learning Research (TMLR) arXiv, reviewed on OpenReview.

• Selim Kuzucu, Jiaee Cheong, Hatice Gunes, Sinan Kalkan. Uncertainty as a Fairness Measure. Published at the ACM Journal of Artificial Intelligence Research (JAIR). JAIR Journal Page.

• Jiaee Cheong, Selim Kuzucu, Sinan Kalkan, Hatice Gunes. Towards Gender Fairness for Mental Health Prediction. Published at IJCAI 2023 AI and Social Good Track (Aug'2023). IJCAI Conference Page.

Education

ELLIS Program - Max Planck Institute for Informatics

ELLIS PhD - Industry Track with Google Zurich

Middle East Technical University

Bachelor of Science - Computer Engineering

High Honor Student with GPA: 3.66/4.00. Graduated in Top 5% (14/305) of the class

Community Service

• Reviewer for NeurIPS and ICML since 2024, ICLR since 2025, IEEE T-PAMI since 2025 and WACV since 2024.

Honors, Awards and Scores

• June, 2022 First place in Guided Research Symposium at METU Computer Engineering Department out of 30 different undergraduate research projects with my work in developing a novel one-pass uncertainty quantification method

• August, 2021 Scored 118/120 in TOEFL (R: 30, L: 29, S: 30, W: 29)

• September, 2019 Earned METU Development Foundation’s Academic Merit Scholarship for ranking in top 0.01% (270th out of 2.5 million takers) in Turkish University Entrance Exam

• March, 2018 Won FIRST Robotics Competition Shanghai Regionals Rookie Allstar Award out of about 50 other rookie teams as the founder team captain of Turkish province of Aydin’s first competitive robotics team