You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
My feeling is that the tasks we currently focus on are simply too easy/don't require it. If imagenet can be solved on texture/local features alone, why bother using compositionality?
because it is not working for language understanding, flexible reasoning, high-level planning, etc. imagenet addresses just a tiny slice of general intelligence.
To be fair, ImageNet is kind of an old dataset. Newer ones In NLP (http://ai.stanford.edu/blog/beyond-local-pattern-matching/ …) as well as CV (https://visualcommonsense.com/ ) are far closer to needing common sense. tbh people have simply not tried to address QA of the difficulty posed in article, as far as I can tell...
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
