I was about to say the same (Gary Marcus’ substack here).
In defense of Marcus, he often complains about AI companies refusing to give him access to their newer models. If your language/​image model is really as awesome as advertised, surviving the close scrutiny of a skeptical scientist should not be a problem, but apparently it is.
I was about to say the same (Gary Marcus’ substack here).
In defense of Marcus, he often complains about AI companies refusing to give him access to their newer models. If your language/​image model is really as awesome as advertised, surviving the close scrutiny of a skeptical scientist should not be a problem, but apparently it is.