Fawaz Sammani

I'm a 3rd year PhD student at the ETRO department at Vrije Universiteit Brussel (VUB), supervised by Prof. Nikos Deligiannis, under the FWO Fellowship Grant. I was also a visiting researcher at the Max Planck Institute for Informatics in Saarbrücken, Germany supervised by Dr. Jonas Fischer. I did my Masters degree in Applied Computer Science at the VUB under the VUB Masters Scholarship Programme. My research interests focus on vision-language, zero-shot learning, generative models and inference-time approaches.

Email / Scholar / X / Github / LinkedIn

Publications

	Temporal Concept Dynamics in Diffusion Models via Prompt-Conditioned Interventions Ada Görgün, Fawaz Sammani, Nikos Deligiannis, Bernt Schiele, Jonas Fischer (equal contribution) ICLR 2026* Paper \| Code \| Website
	Zero-Shot Natural Language Explanations Fawaz Sammani, Nikos Deligiannis ICLR 2025 Paper
	Interpreting and Analysing CLIP's Zero-Shot Image Classification via Mutual Knowledge Fawaz Sammani, Nikos Deligiannis NeurIPS 2024 Paper \| Code
	CLIP-Free, Label-Free, Zero-Shot Concept Bottleneck Models Fawaz Sammani, Jonas Fischer, Nikos Deligiannis arXiv 2025 Paper
	Uni-NLX: Unifying Textual Explanations for Vision and Vision-Language Tasks Fawaz Sammani, Nikos Deligiannis ICCV Workshops 2023 (Honorable Mention Award) Paper \| Code \| Talk
	Visualizing and Understanding Contrastive Learning Fawaz Sammani, Boris Joukovsky, Nikos Deligiannis IEEE Transactions on Image Processing 2023 Paper \| Code
	Model-Agnostic Visual Explanations via Approximate Bilinear Models Boris Joukovsky, Fawaz Sammani, Nikos Deligiannis ICIP 2023 Paper
	NLX-GPT: A Model for Natural Language Explanations in Vision and Vision-Language Tasks Fawaz Sammani, Tanmoy Mukherjee, Nikos Deligiannis CVPR 2022 (Oral) Paper \| Code
	Show, Edit and Tell: A Framework for Editing Image Captions Fawaz Sammani, Luke Melas-Kyriazi CVPR 2020 Paper \| Code
	Look and Modify: Modification Networks for Image Captioning Fawaz Sammani, Mahmoud Elsayed BMVC 2019 Paper \| Code

Academic Service

Reviewer: I served as a reviewer for the following venues (most recent first): CVPR 2026, ICLR 2026, NeurIPS 2025, ICML 2025, CVPR 2025, ICLR 2025, NeurIPS 2024, ECCV 2024, ICCV 2023, BMVC 2024, BMVC 2023. I also review regularly for Transactions on Machine Learning Research (TMLR)

Teaching: Teaching assistant (teaching practical sessions, assigning and correcting projects, creating theory and practical material) for the Deep Learning course, Masters of Applied Computer Science (MACS), Vrije Universiteit Brussel, for Academic Years 2022-2023, 2023-2024, 2024-2025, 2025-2026

Supervision

I have supervised the following Master Thesis projects:

Natural Language Explanations for Medical Imaging, by Tolga Dinc
Parameter-Efficient Inference-Time Fine-Tuning of GPT-2: Exploring Feature Inversion to Privacy Attacks, by Alex-Mihai Lapusan
Vizualizing and Understanding Semi-Supervised Learning, by Malek Boudiaf

Summer Schools

Attended the International Computer Vision Summer School (ICVSS) 2023, Sicily, Italy