Yasaman Razeghi

Download my resume!

Besides that, I enjoy gardening, reading, and writing short stories.

About me

I'm Yasaman, a Ph.D. student at the University of Calirofnia, Irvine working with Prof. Sameer Singh. My research interests lie primarily in Natural Language Processing. I'm interested in using language models under few-shots settings to solve common-sense reasoning problems. I am currently a student researcher at DeepMind.

Publications

Snoopy: An Online Interface for Exploring the Effect of Pretraining Term Frequencies on Few-Shot LM Performance

Published in Proceedings of the The 2022 Conference on Empirical Methods in Natural Language Processing: System Demonstrations

Authors: Yasaman Razeghi, Raja Sekhar Reddy Mekala, Robert L Logan Iv, Matt Gardner, Sameer Singh

A theoretically grounded benchmark for evaluating machine commonsense

Published in arXiv

Authors: Henrique Santos, Ke Shen, Alice M Mulvehill, Yasaman Razeghi, Deborah L McGuinness, Mayank Kejriwal

Impact of Pretraining Term Frequencies on Few-Shot Reasoning

Published in arXiv

Authors: Yasaman Razeghi, Robert L. Logan IV, Matt Gardner, Sameer Singh

Deep Bucket Elimination

Published in IJCAI 2021, Proceedings of the International Joint Conferences on Artificial Intelligence, August 2021

Authors: Yasaman Razeghi, Kalev Kask, Yadong Lu, P. Baldi, Sakshi Agarwal, and R. Dechter

Deriving Behavioral Tests from Common Sense Knowledge Graphs

Published in CSKG 2021, AAAI Workshop on Common Sense Knowledge Graphs, February 2021

Authors: Yasaman Razeghi, Robert L. Logan IV, and Sameer Singh

AutoPrompt: Eliciting Knowledge from Language Models with Automatically Generated Prompts

Published in EMNLP 2020, Empirical Methods in Natural Language Processing, November 2020

Authors: Taylor Shin*, Yasaman Razeghi*, Robert L. Logan IV*, Eric Wallace, and Sameer Singh