Taha Koleilat

PhD Student, Mila – Quebec AI Institute & Concordia University

profile_img.jpg

Montreal, Canada

Hi, I am Taha Koleilat, a PhD student in Electrical & Computer Engineering at Concordia University and Mila – Quebec AI Institute, supervised by Prof. Hassan Rivaz and Prof. Yiming Xiao at the IMPACT and Health-X Labs.

My research lies at the intersection of medical image analysis, vision–language models, and foundation models. I develop data-efficient, generalizable, and interpretable multimodal learning methods that bridge medical images and clinical text, with a focus on robustness in low-data, cross-domain, and clinically realistic settings. Currently, I am working on text-driven medical image segmentation, vision–language model adaptation, and uncertainty-aware multimodal learning.

My work has been published at venues including CVPR, MICCAI, TMLR, and Medical Image Analysis, and is supported by the FRQNT Doctoral Scholarship. I serve as a reviewer for IEEE TPAMI, IEEE TMI, MICCAI, and MedIA. Previously, I was an Applied Scientist Intern at Amazon, working on computer vision research with the Ring team. Before my PhD, I received my BEng in Computer & Communications Engineering from the American University of Beirut with High Distinction (4.0/4.0) and a minor in Economics.

Email / Google Scholar / GitHub / LinkedIn / CV

News

Jun 15, 2026 Started my internship as an Applied Scientist at Amazon, working with the Ring team!
May 10, 2026 Our paper Evi-Steer was early accepted to MICCAI 2026, see you in Strasbourg, France!
Apr 10, 2026 Our paper CLIP-SVD was accepted to Transactions on Machine Learning Research (TMLR). Congrats to all co-authors!
Apr 10, 2026 Received first place award (PhD) at the 10th Graduate Student Research Conference, hosted at Concordia University!
Mar 30, 2026 I am serving as a reviewer for MICCAI 2026.
Feb 21, 2026 Two papers MedCLIPSeg and MedQwen were accepted to CVPR 2026, see you in Denver, Colorado!
Dec 24, 2025 Received the IEEE TMI Gold-Level Distinguished Reviewer Certificate for the second consecutive year.
Oct 5, 2025 Our paper MedViTV2 was accepted to Applied Soft Computing. Congrats to all co-authors!
Jul 30, 2025 Our paper GroundingDINO-US-SAM was accepted to IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control (TUFFC). Congrats to all co-authors!
Jul 22, 2025 Our paper MedCLIP-SAMv2 was accepted to Medical Image Analysis (MedIA). Congrats to all co-authors!
Jul 10, 2025 Our paper TextSAM-EUS was accepted to the CVAMD Workshop (ICCV 2025), see you in Honolulu, Hawaii!
Apr 30, 2025 I was awarded the FRQNT Doctoral Training Scholarship, ranked 2nd province-wide, supporting my research on Vision–Language Models for healthcare.
Mar 15, 2025 I am serving as a reviewer for MICCAI 2025.
Feb 27, 2025 Our paper BiomedCoOp was accepted to CVPR 2025, see you in Nashville, Tennessee!
Jan 25, 2025 I received the Concordia University Doctoral Graduate Fellowship, awarded to exceptional doctoral students for academic and research excellence.
Nov 20, 2024 I received the IEEE TMI Gold-Level Distinguished Reviewer Certificate in recognition of my reviewing contributions.
Jul 1, 2024 Our paper MedCLIP-SAM was accepted to MICCAI 2024, see you in Marrakesh, Morocco!
Jun 15, 2023 I graduated from the American University of Beirut with High Distinction (4.0/4.0 GPA) in Computer & Communications Engineering.
Apr 30, 2023 Our capstone project “COVID-19 Indoor Access Rules Verification using ML” won the Dean’s Award for Creative Achievement, placing 1st among 30 groups.

Selected publications

* denotes joint first authors

2027

  1. evisteer_overview.png
    MICCAI 2027
    Evi-Steer: Learning to Steer Biomedical Vision-Language Models through Efficient and Generalizable Evidential Tuning
    Taha Koleilat, Hassan Rivaz, and Yiming Xiao
    In Medical Image Computing and Computer Assisted Intervention – MICCAI 2026, 2027

2026

  1. medclipseg_overview.png
    CVPR 2026
    MedCLIPSeg: Probabilistic Vision-Language Adaptation for Data-Efficient and Generalizable Medical Image Segmentation
    Taha Koleilat, Hojat Asgariandehkordi, Omid Nejatimanzari, Berardino Barile, Yiming Xiao, and Hassan Rivaz
    In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026
  2. medqwen_overview.jpg
    CVPR 2026
    Sparse Spectral LoRA: Routed Experts for Medical VLMs
    Omid Nejatimanzari, Hojat Asgariandehkordi, Taha Koleilat, Yiming Xiao, and Hassan Rivaz
    In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026
  3. CLIP-SVD.png
    TMLR 2026
    CLIP-SVD: Efficient and Interpretable Vision–Language Adaptation via Singular Values
    Taha Koleilat, Hassan Rivaz, and Yiming Xiao
    Transactions on Machine Learning Research, 2026

2025

  1. biomedcoop_overview.png
    CVPR 2025
    Biomedcoop: Learning to prompt for biomedical vision-language models
    Taha Koleilat, Hojat Asgariandehkordi, Hassan Rivaz, and Yiming Xiao
    In Proceedings of the Computer Vision and Pattern Recognition Conference, 2025
  2. medclipsamv2_overview.png
    MedIA 2025
    Medclip-samv2: Towards universal text-driven medical image segmentation
    Taha Koleilat, Hojat Asgariandehkordi, Hassan Rivaz, and Yiming Xiao
    Medical Image Analysis, 2025
  3. textsam_eus.png
    ICCV CVAMD 2025
    Textsam-eus: Text prompt learning for sam to accurately segment pancreatic tumor in endoscopic ultrasound
    Pascal Spiegler*, Taha Koleilat*, Arash Harirpoush, Corey S Miller, Hassan Rivaz, Marta Kersten-Oertel, and Yiming Xiao
    In Proceedings of the IEEE/CVF International Conference on Computer Vision Workshops (CVAMD), 2025

2024

  1. medclipsam_framework.jpg
    MICCAI 2024
    Medclip-sam: Bridging text and image towards universal medical image segmentation
    Taha Koleilat, Hojat Asgariandehkordi, Hassan Rivaz, and Yiming Xiao
    In International conference on medical image computing and computer-assisted intervention, 2024