
Sarah Wiegreffe
Georgia Insitute of Technology
language models
transformers
commonsense
gpt
model editing
language models; evaluation; question answering; probability
plausibility judgements
5
presentations
3
number of views
SHORT BIO
Sarah is a final-year PhD student in computer science at Georgia Tech, advised by Mark Riedl. Her research is on interpretability of deep learning models for NLP.
Presentations

Increasing Probability Mass on Answer Choices Does Not Always Improve Accuracy | VIDEO
Sarah Wiegreffe and 4 other authors

Editing Common Sense in Transformers | VIDEO
Anshita Gupta and 6 other authors

Measuring Association Between Labels and Free-Text Rationales
Sarah Wiegreffe and 2 other authors

Measuring Association Between Labels and Free-Text Rationales
Sarah Wiegreffe and 2 other authors

Measuring Association Between Labels and Free-Text Rationales
Sarah Wiegreffe and 2 other authors