[kaiser]

There are a number of experiments where a rod of metal is stretched in a machine and the stress and strain recorded. Quite a number can be found on YouTube.
There is always quite a noticeable deformation taking place, before the rod breaks, and there is a noticeable narrowing as well.
The stretch caused by thermal expansion is most likely only a fraction of the stretch caused by the thread, and nowhere near the yield point of the bolt, in the k4.
I think the confusion arises by some people using the term yield as a simple reference to the bolt "giving" and being designed to do so.
The "yield" point is however quite specific in engineering terms, as the point of no return. I think this could be the cause of some misunderstanding.
I have never seen a k4 bolt stretched. I have however seen a few V6 bolts broken, but they are much shorter. I would not re-use a V6 head bolt, but quite happily a k4.