Yizhong Wang
PhD student @ University of Washington
language models
zero-shot learning
benchmark
few-shot learning
knowledge
cross-task generalization
instruction tuning
instruction-based models
temporal alignment
4
presentations
1
number of views
SHORT BIO
Yizhong Wang is a 4th-year PhD student at the University of Washington. His current research focuses on building general-purpose models that can follow instructions and generalize across tasks in a zero-shot or few-shot fashion. He is also broadly interested in question answering and knowledge-augmented language models.
Presentations
Set the Clock: Temporal Alignment of Pretrained Language Models
Bowen Zhao and 4 other authors
Super-NaturalInstructions: Generalization via Declarative Instructions on 1600+ NLP Tasks
Yizhong Wang
Probing Across Time: What Does RoBERTa Know and When?
Zeyu Liu and 1 other author
Probing Across Time: What Does RoBERTa Know and When?
Zeyu Liu and 1 other author