Swabha Swayamdipta

Postdoctoral Investigator • MOSAICAllen Institute for AI

My research focuses on studying biases in datasets and models. Good biases, such as structural inductive biases help language understanding - check out my PhD thesis on these. But biases can be undesirable, e.g. spurious correlations commonly found in crowd-sourced, large-scale datasets due to annotation artifacts, or social prejudices of human annotators and task designers, which are difficult to rid!


I obtained my PhD from Carnegie Mellon University in May 2019, where I was advised by Noah Smith and Chris Dyer. During most of my PhD I was a visiting student at the University of Washington in Seattle. I received a Masters degree from Columbia University where I was advised by Owen Rambow, and my B.Tech in CSE from NIT Calicut.

Update I will be starting as an Assistant Professor in the USC Viterbi Department of CS in Fall 2022!


May 5, 2021 Paper on On-the-Fly Controlled Text Generation with Experts and Anti-Experts to appear at ACL 2021.
Mar 4, 2021 Guest lecture on Transfer Learning at UW Stats: UW DATA 598 Statistical Deep Learning, taught by Zaid Harachoui.
Mar 1, 2021 Check out our new pre-print on contrastive explanations for model decisions! Work with my intern Alon Jacovi and others!
Feb 24, 2021 Talk at the NERT Seminar at Georgetown University! So honored to be an elected speaker :)
Feb 12, 2021 Invited talk at the NLP Seminar at Georgia Tech!
Feb 3, 2021 Check out our new pre-print on an evaluation metric for open-ended text generation, MAUVE !
Feb 1, 2021 New ACL submission on controlled generation, with exciting applications. Keep an eye out!
Jan 11, 2021 Paper on Challenges in Social Bias Mitigation in Hate Speech Detection to appear at EACL 2021!
Dec 3, 2020 Guest lecture in Eunsol Choi’s Topics in NLP class at UT Austin on Biases and Interpretability.
Nov 2, 2020 Was delighted to be an invited speaker for Responsible AI at the Microsoft E+D Product Leaders Conference.