Elnaz Rahmati
PhD Candidate in Computer Science at USC
I am a PhD student in Computer Science at the University of Southern California, advised by Prof. Morteza Dehghani. My research lies at the intersection of natural language processing and cognitive science, with a particular focus on continual learning and multilinguality. More broadly, I study how language models learn, represent, and generalize knowledge. My current and past work spans abstraction in LLMs, multilingual language modeling, conversational modeling, computational social science, and low-resource language technologies.
Publications
2026
The Subjectivity of Respect in Police Traffic Stops: Modeling Community Perspectives in Body-Worn Camera Footage
Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics
Flip-Flop Consistency: Unsupervised Training for Robustness to Prompt Perturbations in LLMs
Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics
Structural Abstraction as an Inductive Bias for Non-Stationary Language Model Training
arXiv preprint arXiv:2603.17198
2025
CoCo-CoLa: Evaluating and Improving Language Adherence in Multilingual LLMs
Proceedings of the 5th Workshop on Multilingual Representation Learning (MRL)
The Moral Foundations Reddit Corpus
Proceedings of the 15th Conference on Language Resources and Evaluation
2024
Evaluating Creativity and Deception in Large Language Models: A Simulation Framework for Multi-Agent Balderdash
Wordplay Workshop at the Association for Computational Linguistics: ACL 2024
GE2PE: Persian End-to-End Grapheme-to-Phoneme Conversion
Findings of the Association for Computational Linguistics: EMNLP 2024, 3426–3436
2022
Naab: A Ready-to-Use Plug-and-Play Corpus for Farsi
Journal of Artificial Intelligence, Applications and Innovations