Learning the Wrong Lessons: Syntactic-Domain Spurious Correlations in Language Models
Type
Publication
The Thirty-ninth Annual Conference on Neural Information Processing Systems
Chantal Shaib
Chantal Shaib is a postdoctoral researcher at MIT. Her research interests are focused on evaluating linguistic variation and the downstream implications for robustness and novelty in language models. Chantal completed her PhD in Computer Science at Northeastern University, where she was awarded the Outstanding PhD Student Award in Research for her work on lexical and syntactic cues in LLM writing. Previously, she completed her MSc in Applied Computing at the University of Toronto, and her BSc in Neuroscience at Carleton University.