Another method, create a formal method for processing complaints from Twitter that diverts the hateful energy from individuals towards a larger decision-making body. Let us say that a random 24 people jury of MIT peers had to decide by secret ballot whether the researcher in question gets approved or disinvited, and that the result is 14:10 to keep him.
The Twitter mob cannot really do anything with such result, if individual voting behavior was anonymous.