Research & Publications

Research Interests

Privacy and Security in Emerging Technologies: Privacy in LLMs and AI Agents; Privacy in XR and Behavioral Data; Privacy-Preserving and Policy-Aware Machine Learning.

Preprints and Under Review

ForgetMeNot: Understanding Memory-Driven Privacy Risks in Consumer AI Agents
Ismat Jarin, Tu Le, and Athina Markopoulou
In preparation.

MotionPrint: An Open-World Framework for Motion-Based User Identification in XR
Raj Mohanty*, Ismat Jarin*, and Athina Markopoulou
Under submission, 2026.

* Equal contribution and co-first authorship.

Peer-Reviewed Publications (Privacy, Security and Safety in Emerging Tech, e.g., AI Agents, LLMs and XR)

Dead Men Tell No Tales: Assessing Post-Mortem Data Protection in GenAI Chatbots
Elina van Kempen*, Ismat Jarin*, and Chloe Georgiou
10th Workshop on Technology and Consumer Protection (ConPro ’26), co-located with IEEE Symposium on Security and Privacy, 2026.
[Paper] [Slides]

* Equal contribution and co-first authorship.

VRProfiLens: User Profiling Risk in Consumer Virtual Reality Apps
Ismat Jarin, Olivia F., Yu Duan, Tu Le, and Athina Markopoulou
16th Usable Security and Privacy Symposium (USEC), co-located with NDSS, 2026.
Poster presented at the WiCyS Annual Conference, Washington, DC, 2026.
[Paper] [Poster] [Code][Talk]

BehaVR: User Identification Based on VR Sensor Data
Ismat Jarin, Yu Duan, Rahmadi Trimananda, Hao Cui, Salma Elmalaki, and Athina Markopoulou
Proceedings on Privacy Enhancing Technologies, 2025, Issue 1.
Poster presented at the UCI AI and Cloud Workshop, 2024.
[Paper] [Poster] [Code][Talk]

MLSherlock: An Audit Lens for Machine Unlearning Systems
Ismat Jarin, Tu Le, and Athina Markopoulou
Women in Machine Learning Workshop (WiML), co-located with NeurIPS, 2024.
[Paper] [Poster] [Code][Talk]

Maverick-Aware Shapley Valuation for Client Selection in Federated Learning
Mengwei Yang, Ismat Jarin, Baturalp Buyukates, Salman Avestimehr, and Athina Markopoulou
ISIT Workshop on Information-Theoretic Methods for Trustworthy Machine Learning, 2024.
[Paper]

Peer-Reviewed Publications (Privacy Preserving Machine Learning)

MIAShield: Defending Membership Inference Attacks via Preemptive Exclusion of Members
Ismat Jarin and Birhanu Eshete
Proceedings on Privacy Enhancing Technologies, 2023.
Co-authored work featured in a DEF CON 30 Biohacking Village talk, Las Vegas, Nevada, 2022.
[Paper] [Code] [Talk]

DP-UTIL: Comprehensive Utility Analysis of Differential Privacy in Machine Learning
Ismat Jarin and Birhanu Eshete
ACM Conference on Data and Application Security and Privacy (CODASPY), 2022.
[Paper] [Code] [Talk]

PRICURE: Privacy-Preserving Collaborative Inference in a Multi-Party Setting
Ismat Jarin and Birhanu Eshete
ACM Workshop on Security and Privacy Analytics (IWSPA), 2021.
[Paper] [Code] [Talk]

Peer-Reviewed Publications (Classical ML, Image and Algorithomic Analysis, Image Encoding)

Retinal Vein Detection Using a Residual Block-Incorporated U-Net Architecture and Fuzzy Inference System
Ismat Jarin, Tahsin Mostafiz, and Celia Shahnaz
IEEE International Women in Engineering Conference on Electrical and Computer Engineering (WIECON-ECE), 2018.
[Paper]

Natural and Medical Image Encryption Using Self-Adaptive Permutation and DNA Encoding
Ismat Jarin and Celia Shahnaz
IEEE International Women in Engineering Conference on Electrical and Computer Engineering (WIECON-ECE), 2018.
[Paper]

Performance Evaluation of MLD and ZFE in a MIMO-OFDM System Impaired by Carrier Frequency Offset, Phase Noise, and Timing Jitter
Ismat Jarin and S. P. Majumdar
IEEE International Conference on Telecommunications and Photonics (ICTP), 2017.
[Paper]

Performance Evaluation of a SISO-OFDM System in the Presence of CFO, Timing Jitter, and Phase Noise for Rayleigh and Rician Fading Channels
Ismat Jarin, R. Sharmin, and S. P. Majumdar
IEEE Region 10 Humanitarian Technology Conference (R10-HTC), 2017, pp. 498–501.
[Paper]

Page updated

Google Sites

Report abuse