Hi, I'm Akhtar

Research Scientist (CV)
MBZUAI, Abu Dhabi
United Arab Emirates

its<dot>akhtar90<at>gmail<dot>com
akhtar<dot>munir<at>mbzuai<dot>ac<dot>ae

I am a Research Scientist in Computer Vision at MBZUAI, Abu Dhabi, working on Reliable Multimodal AI, Foundation Models, Vision-Language Models, Model Calibration, and Agentic AI for geospatial and environmental applications. My work focuses on building AI systems that remain accurate and robust under real-world conditions.

A central theme of my research is Reliability: understanding when models are confident, when they fail, and how they should be evaluated in practice. My work has led to publications at CVPR, ICCV, NeurIPS, ICLR, and TPAMI, including CVPR/ICCV highlights and a Best Paper at ICML TerraBytes. My goal is to translate rigorous machine learning research into reliable AI systems for real-world decision-making.

Advisors: Dr. Salman Khan, Dr. Fahad S Khan.

Publications (Selected)

Ashshak Sharifdeen, Fahad Shamshad, Muhammad Akhtar Munir, Abhishek Basu, Mohamed Insaf Ismithdeen, Jeyapriyan Jeyamohan, Chathurika Sewwandi Silva, Karthik Nandakumar, Muhammad Haris Khan

"Towards Calibrating Prompt Tuning of Vision-Language Models". In CVPR (2026).

[Paper] [Code]

Akashah Shabbir, Muhammad Umer Sheikh, Muhammad Akhtar Munir, Hiyam Debary, Mustansar Fiaz, Muhammad Zaigham Zaheer, Paolo Fraccaro, Fahad Shahbaz Khan, Muhammad Haris Khan, Xiao Xiang Zhu, Salman Khan

"OpenEarthAgent: A Unified Framework for Tool-Augmented Geospatial Agents". In arXiv (2026).

[Paper] [Code]

Akashah Shabbir*, Muhammad Akhtar Munir*, Akshay Dudhane*, Muhammad Umer Sheikh, Muhammad Haris Khan, Paolo Fraccaro, Juan Bernabe Moreno, Fahad Shahbaz Khan, Salman Khan.

"ThinkGeo: Evaluating Tool-Augmented Agents for Remote Sensing Tasks". In arXiv (2026). *Equally contributing first authors.

[Paper] [Code]

Muhamad Sohail Danish, Muhammad Akhtar Munir, Syed Roshaan Ali Shah, Muhammad Haris Khan, Rao Muhammad Anwer, Jorma Laaksonen, Fahad Shahbaz Khan, Salman Khan.

"TerraFM: A Scalable Foundation Model for Unified Multisensor Earth Observation". In ICLR (2026).

[Paper] [Code]

Yohan Abeysinghe, Muhammad Akhtar Munir, Sanoojan Baliah, Ron Sarafian, Fahad Shahbaz Khan, Yinon Rudich, Salman Khan.

"Synergistic Neural Forecasting of Air Pollution with Stochastic Sampling".

[Paper] [Code]

Muhamad Sohail Danish*, Muhammad Akhtar Munir*, Syed Roshaan Ali Shah, Kartik Kuckreja, Fahad Shahbaz Khan, Paolo Fraccaro, Alexandre Lacoste, and Salman Khan.

"GEOBench-VLM: Benchmarking Vision-Language Models for Geospatial Tasks". In ICCV (2025) as Highlight. *Equally contributing first authors.

[Paper] [Code]

Vishal Nedungadi, Muhammad Akhtar Munir, Marc Rußwurm, Ron Sarafian, Ioannis N. Athanasiadis, Yinon Rudich, Fahad Shahbaz Khan, and Salman Khan.

AirCast: Improving Air Pollution Forecasting Through Multi-Variable Data Alignment". In Terrabytes-ICML (2025) as Best Paper.

[Paper] [Code]

Ashshak Sharifdeen, Muhammad Akhtar Munir, Sanoojan Baliah, Salman Khan, and Muhammad Haris Khan.

"O-TPT: Orthogonality Constraints for Calibrating Test-time Prompt Tuning in Vision-Language Models". In CVPR (2025) as Highlight.

[Paper] [Code]

Sagar Soni, Akshay Dudhane, Hiyam Debary, Mustansar Fiaz, Muhammad Akhtar Munir, Muhammad Sohail Danish, Paolo Fraccaro, Campbell D Watson, Levente J Klein, Fahad Shahbaz Khan, Salman Khan.

"EarthDial: Turning Multi-sensory Earth Observations to Interactive Dialogues". In CVPR (2025).

[Paper] [Code]

Muhammad Akhtar Munir, Fahad Shahbaz Khan, and Salman Khan.

"Efficient Localized Adaptation of Neural Weather Forecasting: A Case Study in the MENA Region". In CCAI-NeurIPS (2024).

[Paper] [Code]

Sohail Danish, Muhammad Haris Khan, Muhammad Akhtar Munir, Saquib Sarfraz, and Mohsen Ali.

"Improving Single Domain-Generalized Object Detection: A Focus on Diversification and Alignment". In CVPR (2024).

[Paper] [Code]

Muhammad Akhtar Munir, Salman Khan, Muhammad Haris Khan, Mohsen Ali, and Fahad Shahbaz Khan.

"Cal-DETR: Calibrated Detection Transformer". In NeurIPS (2023).

[Paper] [Code]

Muhammad Akhtar Munir, Muhammad Haris Khan, Muhammad Saquib Sarfraz, and Mohsen Ali.

"Domain Adaptive Object Detection via Balancing between Self-Training and Adversarial Learning". In IEEE TPAMI (2023).

[Paper] [Project Page]

Muhammad Akhtar Munir, Muhammad Haris Khan, Salman Khan, and Fahad Shahbaz Khan.

"Bridging Precision and Confidence: A Train-Time Loss for Calibrating Object Detection". In CVPR (2023).

[Paper] [Code]

Muhammad Akhtar Munir, Muhammad Haris Khan, Muhammad Saquib Sarfraz, and Mohsen Ali.

"Towards Improving Calibration in Object Detection under Domain Shift". In NeurIPS (2022).

[Paper] [Code]

Muhammad Akhtar Munir, Muhammad Haris Khan, Muhammad Saquib Sarfraz, and Mohsen Ali.

"SSAL: Synergizing between Self-Training and Adversarial Learning for Domain Adaptive Object Detection". In NeurIPS (2021).

[Paper] [Project Page]

People

Grateful for the opportunity to work with brilliant minds in recent years, including my advisors {Salman Khan (MBZUAI-UAE, ANU-Australia), Haris Khan (MBZUAI-UAE), Saquib Sarfraz (KIT, Mercedes-Germany), Fahad Khan (MBZUAI-UAE, Linkoping-Sweden), and Mohsen Ali (ITU-Pakistan)}, collaborators, and students. Below is a non-exhaustive list of students and early-career collaborators with whom I have had the privilege to supervise/collaborate, both in the past and currently.

Muhammad Sohail Danish

Ph.D. student at MBZUAI. We are currently collaborating on remote sensing benchmarking and foundation models. He has co-authored papers at ICCV (Highlight) and CVPR as part of our joint research efforts.

Akashah Shabbir

Ph.D. student at MBZUAI. Currently collaborating with her on developing an agentic pipeline for remote sensing tasks.

Vishal Nedungadi

Former researcher at MBZUAI. Collaborated with him on air pollution forecasting, which led to a Best Paper award at the ICML TerraBytes workshop. He is now a Ph.D. student at Wageningen University and Research.

Ashshak Sharifdeen

Researcher at MBZUAI. Currently collaborating with him on model calibration for vision-language models. He co-authored a Highlight paper with us at CVPR.

Yohan Abeysinghe

Researcher at MBZUAI. Currently collaborating on air pollution forecasting using a refined modular framework that integrates diffusion mechanisms for climate modeling.

Syed Roshaan Ali

Currently collaborating on remote sensing benchmarking and foundation models. He contributed to our ICCV paper, which was accepted as a Highlight.

Muhammad Umer

Researcher at MBZUAI. Currently collaborating with him on developing an agentic pipeline for remote sensing tasks.

Services & Honors

Reviewer: CVPR 2023-2026, ICLR 2024-2026, ICCV 2023-2025, NeurIPS 2023-2026, ECCV 2026
Outstanding Reviewer: Recognized as Outstanding Reviewer for CVPR 2025
Awardee: NeurIPS 2022 & 2023 Scholar Award, CVPR 2023 DEI Award
Scholarship: PhD Fellowship, from Information Technology University
Conferences (Attended): CVPR 2023-2025, NeurIPS 2023, BMVC 2024, ICCV 2025

Curriculum Vitae (CV)

Download CV as PDF: Download CV