Research

"The body of work is not necessarily something one looks back on to feel good about oneself, but rather, it all informs what you do next and how you do it."
- Arundhati Roy

These days, I am mainly working on the topic of the reliability of multimodal models. During my PhD, I have also worked on the problems of model personalization and label deficiency in distributed learning for computer vision applications.

Recent Publications:

HARMONY: Hidden Activation Representations and Model Output-Aware Uncertainty Estimation for Vision-Language Models
4th Workshop on Uncertainty Quantification for Computer Vision, CVPR workshop 2025
Erum Mushtaq, Zalan Fabian, Yavuz Faruk Bakman, Anil Ramakrishna, Mahdi Soltanolkotabi, Salman Avestimehr

DETECTING UNRELIABLE RESPONSES IN GENERATIVE VISION-LANGUAGE MODELS VIA VISUAL UNCERTAINTY
Quantify Uncertainty and Hallucination in Foundation Models: The next Frontier in Reliable AI ICLR workshop, 2025
Kiana Avestimhehr, Emily Aye, Zalan Mark Fabian, Erum Mushtaq

CroMo-Mixup: Augmenting Cross-Model Representations for Continual Self-Supervised Learning, ECCV 2024,
[Best poster presentation award at USC + Amazon Symposium]
Erum Mushtaq, Duygu Nur Yaldiz, Yavuz Faruk Bakman, Jie Ding, Chenyang Tao, Dimitrios Dimitriadis, Salman Avestimehr

Distributed Architecture Search Over Heterogeneous Distributions,
TMLR 2023
Erum Mushtaq, Chaoyang He, Jie Ding, Salman Avestimehr

Federated alternate training (fat): Leveraging unannotated data silos in federated segmentation for medical imaging, ISBI 2023
Erum Mushtaq, Yavuz Bakman, Jie Ding, Salman Avestimehr

What if kidney tumor segmentation challenge (KiTS19) never happened, ICMLA, 2023
Erum Mushtaq, Jie Ding, Salman Avestimehr

Federated Learning for Clients’ Data Privacy Assurance in Food Service Industry, Applied Sciences 2023
Hamed Taheri Gorji, Mahdi Saeedi, Erum Mushtaq, Hossein Kashani Zadeh, Kaylee Husarik, Seyed Mojtaba Shahabi, Jianwei Qin, Diane E Chan, Insuck Baek, Moon S Kim, Alireza Akhbardeh, Stanislav Sokolov, Salman Avestimehr, Nicholas MacKinnon, Fartash Vasefi, Kouhyar Tavakolian

Ssfl: Tackling label deficiency in federated learning via personalized self-supervision, FL-AAAI 2022 workshop, Best Paper Award
Chaoyang He, Zhengyu Yang, Erum Mushtaq, Sunwoo Lee, Mahdi Soltanolkotabi, Salman Avestimehr

FLamby: Datasets and Benchmarks for Cross-Silo Federated Learning in Realistic Healthcare Settings, NeurIPS 2022
Jean Ogier du Terrail, Samy-Safwan Ayed, Edwige Cyffers, Felix Grimberg, Chaoyang He, Regis Loeb, Paul Mangold, Tanguy Marchand, Othmane Marfoq, Erum Mushtaq, Boris Muzellec, Constantin Philippenko, Santiago Silva, Maria Teleńczuk, Shadi Albarqouni, Salman Avestimehr, Aurélien Bellet, Aymeric Dieuleveut, Martin Jaggi, Sai Praneeth Karimireddy, Marco Lorenzi, Giovanni Neglia, Marc Tommasi, Mathieu Andreux

Complete list of publications: Google Scholar Profile.