They are good at external scientific evaluations, and regularly ace them. Culturally, that's the only thing that matters to this institution.
They do not have a non scientific supervisory board, they don't think they need that, because it's all about the science. What you call abuse, they call dedication to the cause of advancing human knowledge.
However, this type of reporting is extremely dangerous to them. One of the most valuable resources to them is highly skilled, motivated and driven applicants for positions at all levels.
The more this gets out into the light, the more they will need to build the organizational culture to actually do something real about it.
That said, so far, these things are very easy for them to wait out. Very few victims speak out, because either that puts an end to their career, or they are happy to have put that time long behind themselves.
When evaluating whether an institution is accountable, a good default question to ask is, "Is power plural?" In the terminology of the American political order, this is called checks and balances. It's not perfect, but a system of overlapping institutions, whose members are chosen by a plurality of methods and from a plurality of backgrounds, and which have oversight over each other in a loop, are more accountable than unitary institutions.
I'm sure some have attempted to answer this analytically, basically making a "directed power graph" to measure how plural power is, and then correlating that with measures of accountability such as corruption perceptions. This is a huge topic and the second paragraph is my opinion, but that's because I think that's what such an analysis would show.
Daylight is the best disinfectant -- having goals, non-goals, budgets, and expenses as publicly auditable data is a good place to start.
Going deeper, I've had this notion of making a hybrid communication/documentation tool that embeds micro contracts that can be audited. Easily solved by ye olde HN simple weekend project ;-)
I'm interested in how to bake accountability into an organization. I don't like the idea of using whistleblowing as a crutch because things have to get really bad for someone to blow the whistle.