Hi, I'm Alexis! I am a predoctoral researcher (PYI) on the AllenNLP team at the Allen Institute for AI (AI2) working with Matthew Peters and Ana Marasovic. I am broadly interested in building natural language processing systems that can understand and reason about language in human-like ways, and I currently work on methods for improving and evaluating systems' robustness. I am also a frequent co-host of the NLP Highlights Podcast, along with other members of AllenNLP.

I graduated from Harvard in 2020 with a B.A. in Computer Science and Philosophy. During my undergrad, I wrote a joint senior thesis on causal explanations of machine learning models, advised by Hima Lakkaraju and Bernhard Nickel. I have had the privilege of interning at Microsoft Research, where I worked with Tristan Naumann and the rest of the Project Hanover team on biomedical machine reading for precision medicine. I have also been fortunate to have worked closely with Ellie Pavlick and the rest of the Sentence Representation Learning Team at JSALT 2018.

My email is alexisro [at] mit [dot] edu. If you have questions about my work, are interested in collaborating, or want to discuss other research/academia-related topics, please feel free to reach out!

Fall 2021 Update: I am applying to PhD programs.


News

Feb 2022 The third episode in our PhD application series is released.
Oct 2021 NLP Highlights Podcast: Our series on applying to PhD programs is live. The first two episodes are on preparing application materials and applying to programs in Europe vs. the US.
Sept 2021 Paper on learning models for actionable recourse accepted to Neurips 2021.
July 2021 New preprint on generating and perturbing text with semantic controls.