I'm a 2nd year PhD student in Computer Science at
UC San Diego,
fortunate to be advised by Professor
Julian McAuley.
My research interests lie in the intersection of machine learning and personalization.
More specifically, I've been exploring problems related to recommender systems and conversational agents.
Currently, I'm Currently an applied Scientist intern at
Amazon, working on large language models and recommendation systems.
I completed my Master's in Electrical and Computer Engineering at
Purdue University, where I worked with Professor
Qi Guo on a computer vision problem
[arXiv '22]. Prior to my PhD, I was a Signal Processing Engineer at
PSDSARC, developing signal processing systems. I did my undergraduate at
KSU where I worked with Professors
Saleh Alshebeili and
Omar Aldayel on signal processing and convex optimization
[RadarConf '20].
Our framework fuses large language models with collaborative filtering by leveraging conversational context and user histories to improve recommendation accuracy using both short-term and long-term preferences.
@inproceedings{yoon24forecasting,
title = "Forecasting live chat intent from browsing history",
author = "Se-eun Yoon and Ahmad Bin Rabiah and Zaid Alibadi and Surya Kallumadi and Julian McAuley",
year = "2024",
booktitle = "Proceedings of the 28th ACM International Conference on Information and Knowledge Management (CIKM)"
}
We predict user intent from browsing history using a two-stage approach: classify browsing history into high-level categories, then generate detailed intents with a large language model. This method significantly improves performance.
@article{rabiah2022stereoisp,
title={StereoISP: Rethinking Image Signal Processing for Dual Camera Systems},
author={Bin Rabiah, Ahmad and Guo, Qi},
journal={arXiv preprint arXiv:2211.07390},
year={2022}
}
StereoISP is a framework that elevates RGB image reconstruction by leveraging raw measurements from stereo cameras and disparity estimates to improve image quality.
@inproceedings{bin2021haiku,
title={Haiku: Efficient Authenticated Key Agreement with Strong Security Guarantees for IoT},
author={Bin Rabiah, Abdulrahman and Ramakrishnan, KK and Richelson, Silas and Bin Rabiah, Ahmad and Liri, Elizabeth and Kar, Koushik},
booktitle={Proceedings of the 22nd International Conference on Distributed Computing and Networking},
pages={196--205},
year={2021}
}
Haiku is a lightweight IoT protocol ensuring secure communication with periodic session key changes, combining symmetric and public-key cryptography for authentication and key exchange without the need for a trusted third party.
@inproceedings{rabiah2020sdr,
title={SDR-Based Hardware Implementation and Performance Measurement of Transmit Beampattern Design Algorithms},
author={Bin Rabiah, Ahmad and Alsakabi, Mohammed and Aldayel, Omar and Alshebeili, Saleh},
booktitle={2020 IEEE Radar Conference (RadarConf20)},
pages={1--6},
year={2020},
organization={IEEE}
}
We implement and evaluate three transmit beampattern design methods for radar systems, addressing hardware challenges and demonstrating effectiveness using commercial off-the-shelf (COTS) equipment.
