How Good Is This Model Really?

It's very useful to be able to identify a very basic limitation like this. It leads us to really need to wonder, how good is this model really? If we can't find limitations to the model, has it really solved natural language processing? Has it really reached superhuman level language understanding? I know some people that i would, for example, hold the belief that train bort on lots and lots of naturally occurring text and it will learn world knowledge. But i think it still interesting to try and apply these cutting edge techniques to problems even if they seem unlikely to succeed.