Testing learning hypotheses using neural networks by manipulating learning data

Cara Su-Yi Leong and Tal Linzen

Preprint

Prompting sometimes invokes expert-like downward shifts in GPT-4V’s conceptual hierarchies

Cara Su-Yi Leong and Brenden Lake

Paper ⋅ Poster

Language Models Can Learn Exceptions to Syntactic Rules

Cara Su-Yi Leong and Tal Linzen

Paper ⋅ Poster

Long-Distance Dependencies in Continuation Grammar

Cara Su-Yi Leong and Michael Yoshitaka Erlewine

Paper ⋅ Slides ⋅ Handout