On the Fundamental Limits of LLMs at Scale
Published in Under Review, 2025
We take a deep dive into the theoretical limits of Large Language Models (LLMs)
Selected Publications
Please see the full list of my publications on my Google Scholar profile.
Revisiting Near/Remote Sensing with Geospatial Attention
Published in The IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2022, 2022
We propose a novel architecture for the task of overhead image segmentation when auxiliary ground-level images are available
Generative Appearance Flow: A Hybrid Approach for Outdoor View Synthesis
Published in The British Machine Vision Conference (BMVC), 2020
We propose a novel architecture that includes flow-based and direct synthesis sub-networks for novel view synthesis.
Single Image Cloud Detection via Multi-Image Fusion
Published in IEEE International Geosciences and Remote Sensing Symposium (IGARSS), 2020
We present a weakly supervised method for cloud detection based on multi-image fusion
Weakly Supervised Fusion of Multiple Overhead Images
Published in IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), EarthVision, 2019
This paper proposes an image fusion method to combine multiple overhead images to make a single good looking image. Our method is weakly supervised; it learns to removes clouds and other artifacts from images with limited labels.
Weakly Supervised Building Segmentation from Aerial Images
Published in EEE International Geosciences and Remote Sensing Symposium (IGARSS), 2019
This paper proposes a weakly supervised method for building segmentation from overhead images.