Text classification with Transformers is hard. Humans are expensive.
My guess is also that the link got flagged because of who shared it, but that's just a guess.
My guess is also that the link got flagged because of who shared it, but that's just a guess.
My guess would be that you could also figure out word salad to throw at their False Information model that would create false positives.
Orange man rushed out a shot. It must be bad.
Gray man said get the shot. It’s good.
I’m not saying that kind of “thinking” is valid, but it’s apparently widely accepted.
(I got the shot, but I can see why some people would have considerations)