Hi! I'm a Master's student at Stanford University pursuing an M.S. in Computer Science. I currently work on robust and trustworthy machine learning at the Stanford IRIS lab, advised by Professor Chelsea Finn. Recently, I've worked on adaptation of foundation models and out-of-distribution generalization. I completed my undergraduate studies in math (B.S.), also at Stanford. Previously, I worked on math research at the University of Virginia Research Experience for Undergraduates (REU), advised by Professors Charlotte Ure and Ken Ono. I also interned at Snap and Meta, where I worked on weak semi-supervised object detection and infrastructure for privacy-preserving machine learning, respectively.
At Stanford, I served as Professional Development Chair of Tau Beta Pi (Engineering Honors Society).
Please get in touch with me via email if you would like to chat about research!
Publications & Preprints
- AutoFT: Robust Fine-Tuning by Optimizing Hyperparameters on OOD Data
Caroline Choi*, Yoonho Lee*, Annie Chen, Allan Zhou, Aditi Raghunathan, Chelsea Finn. NeurIPS DistShift Workshop, 2023.
- Conservative Prediction via Data-Driven Confidence Minimization
Caroline Choi*, Fahim Tajwar*, Yoonho Lee*, Ananya Kumar, Huaxiu Yao, Chelsea Finn. ICLR TrustML & ME-FOMO Workshops, 2023.
paper | code - Wild-Time: A Benchmark of In-the-Wild Distribution Shift Over Time
Huaxiu Yao*, Caroline Choi*, Bochuan Cao, Yoonho Lee, Pang Wei Koh, Chelsea Finn. NeurIPS (Datasets and Benchmarks Track), 2022.
paper | code | website - Symbol Length in Brauer Groups of Elliptic Curves
Mateo Attanasio*, Caroline Choi*, Andrei Mandelshtam*. Proceedings of the American Mathematical Society, 2023.
paper - On Class Numbers, Torsion Subgroups, and Quadratic Twists of Elliptic Curves
Talia Blum*, Caroline Choi*, Alexandra Hoey*, Jonas Iskander*, Kaya Lakein*, Thomas Martinez*. Transactions of the American Mathematical Society, 2022.
paper | code - On Permutation Weights and q-Eulerian Polynomials
Aman Agrawal*, Caroline Choi*, Nathan Sun*. Annals of Combinatorics, 2020.
paper | code
* denotes equal contribution