Hi! I am a second-year PhD student at Berkeley advised by Dan Klein and Dawn Song. I work on Machine Learning and Natural Language Processing as part of Berkeley AI Research (BAIR), with affiliations in Berkeley NLP, Berkeley Security, and the RISE Lab.
I did my undergrad at the University of Maryland, where I worked with Jordan Boyd-Graber. In 2019, I interned at AI2 with Matt Gardner and Sameer Singh.
If you are an undergrad looking to get started in research, looking for advice on PhD applications, etc., please feel free to email me—I'd love to chat! I am also excited to talk to other researchers and graduate students who share similar research interests.
Robustness We study how an adversary or a distribution shift can affect model behavior. We have identified insightful model errors by appending universal triggers, reducing inputs, and manually editing inputs. Our current research looks to defend real-world systems and also studies threats such as model stealing and data poisoning.
Interpretability We have analyzed the limitations of interpretation methods and helped to facilitate interpretability research with an open-source toolkit and an EMNLP tutorial. Our current research probes pretrained models and studies how training examples affect test predictions.
Dataset Quality We study how "bad data" can lead to undesirable model behavior. We have shown how private and copyright training data can be leaked by language models. We have also identified issues in annotation schemes for NLP datasets. Our current research creates better evaluation data and studies worst-case training examples.