Kuleen Sasse (/kuˈliːn sæs/)

Email: ksasse1[at]jh[dot]edu

Hi! I’m Kuleen Sasse, a Ph.D. student in Computer Science at Johns Hopkins University advised by Gillian Hadfield. I am broadly interested in AI Safety and Responsible AI. I am fortunate for my work to be supported by NSF Graduate Research Fellowship.

Please contact me if you’re interested in collaboration, advice, or just want to chat please reach out to me via my email!

During my undergraduate, I was advised by John Osborne and Mark Dredze and spent summers at the Johns Hopkins Applied Physics Lab as a Research Scientist and Research Engineering intern. Through those 4 years: I worked with a host of people on a diverse set of problems spanning:

Multimodal and Vision-Language Models
Biomedical and Clinical NLP
AI Safety and Fairness
Building Open Pretraining Corpora

In my free time, I like baking, hunting Wikipedia for fun facts, learning about retro phones and computers, geeking about transit, wandering museums, and exploring random and off the beaten path places!

news

Sep 01, 2025	Sparse Autoencoder Features for Classifications and Transferability accepted to EMNLP 2025 Main!
Aug 30, 2025	Released POMELO an independent research project that I worked on at JHU APL that spun off to a full AutoML package!
Jul 30, 2025	Poster Presentation in Vienna at ACL Main 2025! Presenting my work on dog whistles
Apr 08, 2025	Recieved the NSF GRFP! A glimmer of hope during this difficult PhD application season.
Mar 27, 2025	Received Outstanding Senior Award in CS at JHU!

selected publications

Making FETCH! Happen: Finding Emergent Dog Whistles Through Common Habitats

Kuleen Sasse, Carlos Alejandro Aguirre, Isabel Cachola, and 2 more authors

In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Jul 2025

Abs DOI

Dog whistles are coded expressions with dual meanings: one intended for the general public (outgroup) and another that conveys a specific message to an intended audience (ingroup). Often, these expressions are used to convey controversial political opinions while maintaining plausible deniability and slip by content moderation filters. Identification of dog whistles relies on curated lexicons, which have trouble keeping up to date. We introduce FETCH!, a task for finding novel dog whistles in massive social media corpora. We find that state-of-the-art systems fail to achieve meaningful results across three distinct social media case studies. We present EarShot, a strong baseline system that combines the strengths of vector databases and Large Language Models (LLMs) to efficiently and effectively identify new dog whistles.
debiaSAE: Benchmarking and Mitigating Vision-Language Model Bias

Kuleen Sasse, Shan Chen, Jackson Pond, and 2 more authors

Jul 2025
Sparse Autoencoder Features for Classifications and Transferability

Jack Gallifant, Shan Chen, Kuleen Sasse, and 3 more authors

Jul 2025