Talha Ahmed
Bio
I am a Research Assistant at the Algorithms in Theory and Practice (ATP) Lab, Lahore University of Management Sciences (LUMS), working under the supervision of Dr. Hassan Mohy-ud-Din. I graduated with High Distinction in a joint B.S. in Mathematics–Economics with a minor in Computer Science from LUMS in 2024. My research focuses on designing interpretable, robust, and resource-efficient generative models for practical deployment in biomedical and computational imaging settings.
Research Interests
- Efficient & interpretable generative models (diffusion/score-based)
- Medical imaging: segmentation, restoration, reconstruction
- Inverse problems: super-resolution, deblurring, compressed sensing
- Model-Based Deep Learning & unrolled optimization
- Representation learning, robustness & generalization
Academic Journey
- Research Assistant, ATP Lab @ LUMS (Dr. Hassan Mohy-ud-Din) — 2024–Present
- B.S. Mathematics–Economics + Minor CS, LUMS — 2020–2024
News
Wave-GMS accepted to ICASSP 2026! Our lightweight multi-scale generative model for medical image segmentation (~2.6M parameters) achieves state-of-the-art performance and cross-domain generalization. [arXiv] [GitHub]
Submitted Wave-GMS to ICASSP 2026 — a full conference paper on lightweight generative segmentation.
Joined the ATP Lab at LUMS as a Research Assistant under Dr. Hassan Mohy-ud-Din, focusing on efficient generative models for medical imaging and inverse problems.
Successfully graduated from LUMS with a Bachelor of Science in Economics–Mathematics (Joint Major) + Computer Science (Minor).
Awards and Honours
- 2024: Graduated with High Distinction (CGPA ≥ 3.80), LUMS
- 2024: Ranked in top 9% of LUMS SBASSE Batch 2024 (300+ students)
- 2020–2023: Dean’s Honor List (CGPA ≥ 3.60): 2020–21, 2021–22, 2022–23
Contact
I am always happy to connect — whether about research, a shared interest, or just a good conversation. In particular, I welcome:
Research Collaborations
I am open to collaborating on projects at the intersection of generative modelling, medical imaging, and inverse problems — especially work that prioritises efficiency and rigour under real-world constraints.
PhD Opportunities
I am actively seeking PhD positions starting Fall '26 in machine learning, computational imaging, or related areas. If your group works on problems I care about, I would love to hear from you.
Feel free to reach me at talha.123ahmed@live.com · LinkedIn