3 papers
cs.CL2019
Does BERT agree? Evaluating knowledge of structure dependence through agreement relations
Geoff Bacon, Terry Regier
Learning representations that accurately model semantics is an important goal of natural language processing research. Many semantic phenomena depend on syntactic structure. Recent…
cs.CL2018
Efficient human-like semantic representations via the Information Bottleneck principle
Noga Zaslavsky, Charles Kemp, Terry Regier +1
Maintaining efficient semantic representations of the environment is a major challenge both for humans and for machines. While human languages represent useful solutions to this pr…
cs.CL2018
Color naming reflects both perceptual structure and communicative need
Noga Zaslavsky, Charles Kemp, Naftali Tishby +1
Gibson et al. (2017) argued that color naming is shaped by patterns of communicative need. In support of this claim, they showed that color naming systems across languages support…