I'm saying that these quarks are not long-lived, they come and go like other virtual particles. You have this behavior outside of a hadron too, just that you'r

I'm saying that these quarks are not long-lived, they come and go like other virtual particles. You have this behavior outside of a hadron too, just that you're more likely to see a quark-antiquark pair inside a hadron than outside. The sea quarks within the hadron are still virtual particles, though. And often virtual particles aren't considered to be there at all, after all, from that logic one might say that a proton contains all particles (it does, since there always is a possibility of a virtual particle pair of any type springing up). Of course, the possibility of the quarks coming up is more likely since they can be obtained from a gluon, but they're still virtual. We're just dancing around the question "should we count virtual particles when talking about the contents?". The answer usually depends on context. So it's completely ok to say that a proton is just three quarks, since those are the only real particles in it.