>The claims in the document are that MLC shows similar error types to humans, and that this backs some sort of theory of mind The same thing was claimed with l

>The claims in the document are that MLC shows similar error types to humans, and that this backs some sort of theory of mind The same thing was claimed with latent semantic analysis in the late 90s.