aelahi6@uic.edu
Ali Elahi

I am pursuing my Ph.D. at the University of Illinois--Chicago under the guidance of Dr. Cornelia Caragea. I am conducting research in Natural Language Processing (NLP) and Information Retrieval with focus on debate and reasoning of Large Language Models (LLMs). I am also interested in NLP in Finance, and generally solving problems in Finance and Economics that overlap with Data Science.

I spent Sep 2022-Dec 2023 Improving semi-supervised, and domain adaptation algorithms for classification task for predicting water quality in data scares locations. This project was funded by Office of Naval Research and my mentors were Dr. Doina Caragea and Dr. Sam Dorevitch. I am glad that my work took a small step toward enhancing public health and was in line with the United Nations' Sustainable Development Goal of Clean Water and Sanitation.

Beofre 2022 I was a Computer Science student at University of Tehran, and worked as a data scientist at the University of Tehran exploring recommendation algorithms and online and offline evaluations of them. In 2016, I won Iran's National Bronze Medal in Chemistry Olympiad.

Currently Teaching Assistant for Economics and Computation (CS407) taught by Dr. Michael Curry.

Google Scholar: link
research statement: link

Recent Presentations, Teaching, etc.

IEEE BigData Conference: Domain Adaptation in Surface Water Bacteria Level Prediction. [Slides]

CS594: Socially Responsible Language Models; Talk on Bias and Toxicity Mitigation III [Slides]

CS594: Socially Responsible Language Models; Talk on Hallucination Mitigation II [slides]

Teaching Assistant, Data Structure, Fall 2024

CS590: Uncertainty and confidence in Language Models [slides]

Teaching Assistant, Natural Language Processing Spring 2024, Dr. Shweta Yadav,

Publications

Currently working on stance detection and confidence analysis of LLMs in classification tasks. I am trying to redefine concepts such as Data Cartography and Area Under Margin for easier usability with LLMs. Additionally I am working on prompt tuning methods such as dynamic few-shots and sample based prompt generation.

Elahi, A., Shumway, D., Caragea, C., Caragea D., Dorevitch S., (2024, BigData Conference paper). Predicting Surface Water Bacteria Levels Using Transfer Learning and Domain Adaptation [Download]

Elahi, A., Taghvaei, F. (2024, ICAIF AI4Finance poster). Combining Financial Data and News Articles for Stock Price Movement Prediction by Prompting Large Language Models [Poster]

Elahi, A., Taghvaei, F. (2024, BigData  Large Language Models for Finance Workshop paper). Combining Financial Data and News Articles for Stock Price Movement Prediction by Prompting Large Language Models [Download]

Elahi, A., Zirak, A. (2021, Arxive). A Comparative Study of Recommendation Algorithms: Online and Offline Evaluations on a Large-scale Recommender System [Download]

Other Sources

Github

Curriculum vitae: email me and I'll sent it!