Samar Khanna

I am a founding ML/research engineer at a stealth startup led by Stanford Prof. Stefano Ermon, building next-generation large language models.

Recently, I received my MS in CS with a distinction in research from Stanford University, specializing in AI. I worked with Prof. Stefano Ermon on self-supervised learning, generative (diffusion) models, and on building foundation models for geospatial tasks. My research interests center on making ML models generalize effectively to new, real-world data, especially to solve societally relevant tasks such as sustainability.

Before Stanford, I completed my Bachelor’s in CS at Cornell University, where I did research with Dean Kavita Bala and Prof. Bharath Hariharan. I was also actively involved as a team-lead in Cornell Data Science, a student-run project team.

Professionally, I have worked at Aurora as an ML engineer and have completed ML internships at NVIDIA and at Uber ATG.

For fun, I swim and play tennis– to relax, I enjoy reading books and watching movies.

Announcements

Sep 2024	I left Aurora and joined Prof. Ermon to build next-generation, non-autoregressive LLMs at our new startup! Stay tuned for updates!
Jun 2023	I graduated from Stanford!
May 2021	4 years of Cornell all wrapped up…

Selected Publications

ExPLoRA: Parameter-Efficient Extended Pre-Training to Adapt Vision Transformers under Domain Shifts

Samar Khanna, Medhanie Irgau, David B. Lobell, and Stefano Ermon

arXiv preprint arXiv:2406.10973, 2024

arXiv Bib Code Website

@article{explora2024,
  title = {ExPLoRA: Parameter-Efficient Extended Pre-Training to Adapt Vision Transformers under Domain Shifts},
  author = {Khanna, Samar and Irgau, Medhanie and Lobell, David B. and Ermon, Stefano},
  journal = {arXiv preprint arXiv:2406.10973},
  year = {2024},
}

DiffusionSat: A Generative Foundation Model for Satellite Imagery

Samar Khanna, Patrick Liu, Linqi Zhou, Chenlin Meng, Robin Rombach, Marshall Burke, David B. Lobell, and Stefano Ermon

In The Twelfth International Conference on Learning Representations , 2024

arXiv Bib Code Website

@inproceedings{diffusionsat2024,
  title = {DiffusionSat: A Generative Foundation Model for Satellite Imagery},
  author = {Khanna, Samar and Liu, Patrick and Zhou, Linqi and Meng, Chenlin and Rombach, Robin and Burke, Marshall and Lobell, David B. and Ermon, Stefano},
  booktitle = {The Twelfth International Conference on Learning Representations},
  year = {2024},
  url = {https://openreview.net/forum?id=I5webNFDgQ},
}

SatMAE: Pre-training Transformers for Temporal and Multi-Spectral Satellite Imagery

Samar Khanna*, Yezhen Cong*, Chenlin Meng, Patrick Liu, Erik Rozi, Yutong He, Marshall Burke, David B. Lobell, and Stefano Ermon

In Advances in Neural Information Processing Systems , 2022

arXiv Bib Code Website

@inproceedings{satmae2022,
  title = {Sat{MAE}: Pre-training Transformers for Temporal and Multi-Spectral Satellite Imagery},
  author = {Khanna*, Samar and Cong*, Yezhen and Meng, Chenlin and Liu, Patrick and Rozi, Erik and He, Yutong and Burke, Marshall and Lobell, David B. and Ermon, Stefano},
  booktitle = {Advances in Neural Information Processing Systems},
  editor = {Oh, Alice H. and Agarwal, Alekh and Belgrave, Danielle and Cho, Kyunghyun},
  year = {2022},
  url = {https://openreview.net/forum?id=WBhqzpF6KYH},
}

Large Language Models are Geographically Biased

Rohin Manvi, Samar Khanna, Marshall Burke, David Lobell, and Stefano Ermon

In Forty-first International Conference on Machine Learning , 2024

arXiv Bib Code Website

@inproceedings{llmbias2024,
  title = {Large Language Models are Geographically Biased},
  author = {Manvi, Rohin and Khanna, Samar and Burke, Marshall and Lobell, David and Ermon, Stefano},
  booktitle = {Forty-first International Conference on Machine Learning},
  year = {2024},
  url = {https://openreview.net/forum?id=sHtIStlg0v},
}

Denoising Diffusion Bridge Models

Linqi Zhou, Aaron Lou, Samar Khanna, and Stefano Ermon

In The Twelfth International Conference on Learning Representations , 2024

arXiv Bib Code

@inproceedings{ddbm2024,
  title = {Denoising Diffusion Bridge Models},
  author = {Zhou, Linqi and Lou, Aaron and Khanna, Samar and Ermon, Stefano},
  booktitle = {The Twelfth International Conference on Learning Representations},
  year = {2024},
  url = {https://openreview.net/forum?id=FKksTayvGo},
}