Scholar
Muhammad Akhtar Munir
Google Scholar ID: sT-epZAAAAAJ
Mohamed bin Zayed University of Artificial Intelligence, UAE
Citations & Impact
Citations
321
H-index
11
i10-index
11
Publications
19
Co-authors
0
Publications
Resume
Academic Achievements
- ThinkGeo: Evaluating Tool-Augmented Agents for Remote Sensing Tasks (arXiv, 2025)
- TerraFM: A Scalable Foundation Model for Unified Multisensor Earth Observation (arXiv, 2025)
- Synergistic Neural Forecasting of Air Pollution with Stochastic Sampling (Paper Coming Soon)
- GEOBench-VLM: Benchmarking Vision-Language Models for Geospatial Tasks (ICCV, 2025, Highlight)
- AirCast: Improving Air Pollution Forecasting Through Multi-Variable Data Alignment (Terrabytes-ICML, 2025, Best Paper)
- O-TPT: Orthogonality Constraints for Calibrating Test-time Prompt Tuning in Vision-Language Models (CVPR, 2025, Highlight)
- EarthDial: Turning Multi-sensory Earth Observations to Interactive Dialogues (CVPR, 2025)
- Efficient Localized Adaptation of Neural Weather Forecasting: A Case Study in the MENA Region (CCAI-NeurIPS, 2024)
- Improving Single Domain-Generalized Object Detection: A Focus on Diversification and Alignment (CVPR, 2024)
- Cal-DETR: Calibrated Detection Transformer (NeurIPS, 2023)
- Domain Adaptive Object Detection via Balancing between Self-Training and Adversarial Learning (IEEE TPAMI, 2023)
- Bridging Precision and Confidence: A Train-Time Loss for Calibrating Object Detection (CVPR, 2023)
Research Experience
- Serves as a Postdoc. Researcher at MBZUAI, focusing on Computer Vision research. Mainly works on Domain Adaptation, Uncertainty Estimation, and Model Calibration.
Education
- No specific educational background information provided.
Background
- Currently a Postdoctoral Researcher at MBZUAI in Abu Dhabi. Research spans areas like Domain Adaptation, Uncertainty Estimation, Model Calibration, and Vision-Language Models (VLMs). Focused on tackling challenges in Model Calibration, enhancing the accuracy of Weather Modeling, and improving Remote Sensing analysis through VLMs.
Miscellany
- LinkedIn, Google Scholar, GitHub accounts available.