Jaspreet Ranjit


RTH 420

3710 McClintock Ave

Los Angeles, CA 90089

I am a 2nd Year PhD Student at the University of Southern California advised by Prof. Swabha Swayamdipta in the NLP department at the DILL Lab. My research interests lie in investigating societal biases in datasets, measuring the propagation of such biases learned by models in downstream tasks and reasoning about the broader implications of fairness in applications of NLP for social good.

Previously, I was a Research Assistant in the Vision, Language and Learning Lab, working with Prof. Vicente Ordóñez - Rice Univ. and Tianlu Wang - Meta Research on exploring gender biases in Visual Recognition Models. I was also a member of the Link Lab under the mentorship of Prof. Madhur Behl - Univ. of Virginia where I researched methodologies for better traffic scene understanding that contribute to the safety of autonomous vehicles. In my internship at Vimeo, I developed a framework to identify biases in search and recommendation models and prototyped Learning to Rank models for video based search.

I graduated with my Master’s in Computer Science from the University of Virginia in December 2021. I received my Bachelor of Science at the University of Virginia in Computer Science as a Rodman Scholar. Here is a link to my curriculum vitae.


Apr 19, 2024 Received a best poster award at ShowCAIS 2024 for our work: OATH-Frames: Characterizing Online Attitudes Towards Homelessness on Social Media via LLM Assistants. Thank you so much at ShowCAIS!
Apr 3, 2024 Gave a talk at CAIS++ on Characterizing Attitudes Towards Homelessness on Social Media via LLM Assistants
Nov 17, 2023 Characterizing Attitudes Towards Homelessness on Social Media presented @ SoCal NLP Symposium 2023
Oct 31, 2023 Variation of Gender Biases in Visual Recognition Models Before and After Finetuning is accepted to the Algorithmic Fairness through the Lens of Time workshop @ NeurIPS 2023
Aug 22, 2022 Joined as a PhD student at USC, advised by Swabha Swayamdipta