Shawn Im
Hello! I am a PhD student at UW-Madison working with Prof. Sharon Li. I am interested in understanding the behaviors of machine learning models and developing methods that allows these behaviors to be understood by people in hopes of guiding models towards being beneficial for all. Currently, I am working on
- AI Safety
- Learning Theory
- Interpretability
Research
On the Generalization of Preference Learning with DPO
Shawn Im, Yixuan Li
Preprint, 2024
[Paper]
Understanding the Learning Dynamics of Alignment with Human Feedback
Shawn Im, Yixuan Li
In Proceedings of International Conference on Machine Learning (ICML), 2024
Evaluating the Utility of Model Explanations for Model Development
Shawn Im, Jacob Andreas, Yilun Zhou
NeurIPS Workshop on Attributing Model Behavior at Scale (ATTRIB), 2023
[Paper]
Other