Mehak Dhaliwal

Hello! I'm a PhD student in the Department of Computer Science at UC Santa Barbara, where I am advised by Prof. Yao Qin in the REAL AI lab. I'm interested in research in robust and fair AI, as well as AI for healthcare and social good.

Prior to this, I obtained an MS in Computer Science, specialising in AI, from UC San Diego. While there, I researched in Prof. Julian McAuley's lab on data efficient recommendation systems. I obtained my B.Tech. in Computer Science & Engineering from IIT, Delhi, where I worked with Prof. Aaditeshwar Seth on fairness and diversity in content recommendation. I also spent two years as a software engineer in the On-Device AI team at Samsung R&D Institute where I worked on a wide range of projects in Information Retrieval and NLP.

Please feel free to reach out if any of my work or interests align with yours!

News


Publications

NutriBench: A Dataset for Evaluating Large Language Models in Carbohydrate Estimation from Meal Descriptions

Andong Hua*, Mehak Preet Dhaliwal*, Ryan Burke, Yao Qin
Arxiv preprint, 2024

Understanding Hypoglycemia Risk in Unstructured Real-World Physical Activities in Adults with Type 1 Diabetes

Mehak P. Dhaliwal, Kenan Tang, Eleonora M. Aiello, Dessi P. Zaharieva, Rayhan Lal, Cameron Summers, Brandon Arbiter, Kelly Watson, Lauren Figg, Ilenia Balistreri, Ryan S Kingman, Bailey Suh, Yao Qin
American Diabeted Association, 2024

Glycemic Effect of Free-Living Activities in Adults with Type 1 Diabetes

Mehak P. Dhaliwal, Kenan Tang, Eleonora M. Aiello, Yao Qin
American Diabeted Association, 2024

A Shocking Amount of the Web is Machine Translated: Insights from Multi-Way Parallelism

Brian Thompson, Mehak Preet Dhaliwal, Peter Frisch, Tobias Domhan, Marcello Federico
ACL Findings, 2024

Infinite Recommendation Networks: A Data-Centric Approach

Noveen Sachdeva, Mehak Preet Dhaliwal, Carole-Jean Wu, Julian McAuley
Conference on Neural Information Processing Systems (NeurIPS) 2022

Fairness and Diversity in the Recommendation and Ranking of Participatory Media Content

Mehak Preet Dhaliwal*, Muskaan*, Aaditeshwar Seth
International Conference on Advances in Social Networks Analysis and Mining (ASONAM), 2021
Intelligent Information Feed workshop, KDD 2019

On-Device Extractive Text Summarization

Mehak Preet Dhaliwal*, Rishabh Kumar*, Mukund Rungta*, Hemant Tiwari, Vanraj Vala
International Conference on Semantic Computing (ICSC), 2021

Automatic Creation of a Domain Specific Thesaurus Using Siamese Networks

Mehak Preet Dhaliwal, Hemant Tiwari, Vanraj Vala
International Conference on Semantic Computing (ICSC), 2021

TransKP: Transformer based Key-Phrase Extraction

Mukund Rungta*, Rishabh Kumar*, Mehak Preet Dhaliwal*, Hemant Tiwari, Vanraj Vala
International Joint Conference on Neural Networks (IJCNN), 2020

Two-Phase Multimodal Neural Network for App Categorization using APK Resources

Mukund Rungta*, Praneet Prabhakar Sherki*, Mehak Preet Dhaliwal*, Hemant Tiwari, Vanraj Vala
International Conference on Semantic Computing (ICSC), 2020

Education

University of California, Santa Barbara

PhD, Department of Computer Science
September 2023 - Present

University of California, San Diego

Master of Science, Computer Science
September 2021 - June 2023

Indian Institute of Technology, Delhi

Bachelor of Technology (B.Tech.), Computer Science & Eng.
July 2015 - May 2019

Experience

University of California, Santa Barbara

Graduate Student Researcher

Working on AI for healthcare, robustness and fairness in AI, advised by Prof. Yao Qin

September 2023 - Present

University of California, Santa Barbara

Graduate Teaching Assistant

CMPSC 24: Problem Solving with Computers II with Prof. Diba Mirza

September 2023 - December 2023

Amazon Web Services

Applied Scientist Intern

Explored multi-way parallelism in multilingual content on the internet by combining 10.8 billion aligned bi-text pairs across 90+ languages (ccMatrix data set) into a large multi-way parallel data set, called MWccMatrix. Explored various properties of the data including multi-way parallelism for different languages and variations in translation quality, perplexity, and content length across text with varying degrees of multi-way parallelism. [Paper, Code]

June 2023 - September 2023

Tonita

Software Engineer Intern

Developed a question-answering model for handling multiple questions for a context and heterogeneous answer types (extractive, boolean, numerical, no-answer etc.) for reduced inference latency and increased use-case flexibility. Demonstrated an end to end search pipeline utilising developed models for NLP-augmented semantic search.

June 2022 - September 2022

University of California, San Diego

Graduate Teaching Assistant

> CSE 256: Statistical Natural Language Processing with Prof. Ndapa Nakashole, Spring 2023
> DSC 140A: Probabilistic Modeling and Machine Learning with Prof. Justin Eldridge, Winter 2023
> CSE 258: Web Mining and Recommender Systems with Prof. Julian McAuley, Fall 2022
> CSE 256: Statistical Natural Language Processing with Prof. Ndapa Nakashole, Spring 2022

March 2022 - June 2023

Samsung R&D Institute

Senior Software Engineer

Worked in the On-Device AI team on multiple projects in NLP and Information Retrieval including text summarisation, key-phrase extraction, domain-specific semantics, semantic search, app classification and entity extraction.

June 2019 - August 2021

Samsung R&D Institute

Software Research intern

Worked on text classification using Bi-LSTM models

May 2018 - July 2018