-
Assistant Professor, School Of Computing
Publications
- Fateme Hashemi Chaleshtori, Atreya Ghosal & Ana Marasović (2023). On Evaluating Explanation Utility for Human-AI Decision-Making in NLP. XAI in Action: Past, Present, and Future Applications Workshop co-located with NeurIPS 2023.
Published, 12/16/2023.
https://openreview.net/pdf?id=8BR8EaWNTZ
- Jacob K. Johnson & Ana Marasović (2023). How Much Consistency Is Your Accuracy Worth?. The Workshop on Analyzing and Interpreting Neural Networks for NLP (BlackboxNLP) co-located with EMNLP 2023.
Published, 12/07/2023.
https://aclanthology.org/2023.blackboxnlp-1.19/
- Jack Hessel, Ana Marasović, Jena D Hwang, Lillian Lee, Jeff Da, Rowan Zellers, Robert Mankoff & Yejin Choi (2023). Do Androids Laugh at Electric Sheep? Humor “Understanding” Benchmarks from The New Yorker Caption Contest. ACL.
Published, 07/09/2023.
https://aclanthology.org/2023.acl-long.41/
- Alexis Ross, Matthew E. Peters & Ana Marasović (2022). Does Self-Rationalization Improve Robustness to Spurious Correlations. EMNLP 2022.
Published, 12/07/2022.
https://aclanthology.org/2022.emnlp-main.501/
- Shruti Palaskar, Akshita Bhagia, Yonatan Bisk, Florian Metze, Alan W Black & Ana Marasović (2022). On Advances in Text Generation from Images Beyond Captioning: A Case Study in Self-Rationalization. Findings of EMNLP.
Published, 12/07/2022.
https://aclanthology.org/2022.findings-emnlp.194/
- Abhilasha Ravichander, Matt Gardner & Ana Marasović (2022). CondaQA: A Contrastive Reading Comprehension Dataset for Reasoning about Negation. EMNLP.
Published, 12/07/2022.
https://aclanthology.org/2022.emnlp-main.598/
- Ana Marasović, Iz Beltagy, Doug Downey & Matthew E. Peters (2022). Few-Shot Self-Rationalization with Natural Language Prompts. Findings of NAACL.
Published, 07/10/2022.
https://aclanthology.org/2022.findings-naacl.31/
Presentations
- "Challenges in Fostering (Dis)Trust in AI" Keynote at the Workshop on "Trust Perspectives in Machine Learning, Law, and Public Policy'' organized by the Institute for Data, Econometrics, Algorithms, and Learning; located at Northwestern University.
Invited Talk/Keynote,
Presented, 10/01/2023.
- "We propose <insert an explainability method> to increase human trust in AI: What does this mean and how to go about it?" at University of Maryland.
Invited Talk/Keynote,
Presented, 04/01/2023.
- "AI That Masters Language Could Reason about Negation" University of Utah Data Science Seminar.
Invited Talk/Keynote,
Presented, 02/01/2023.