How to dynamically determine the ideal alpha value for linear interpolation

Hey there,

I am linearly interpolating between two poses of a skeleton. For each bone, I use .lerpVectors(a, b, alpha) and .slerpQuaternions(qA, qB, alpha), where alpha is a between [0,1] which basically represents the percentage by which the interpolation has progressed. I update the alpha value in my rendering loop using alpha += 0.5 * dt, where dt is the delta time.

I can also apply an easing function to my alpha value before calling lerp. These factors combined already make the linear interpolation look pretty good. However, I was wondering if there was another approach to use different alpha values for each bone? And how I would I determine the alpha value to use for each bone? An idea I had in mind was to change the alpha value based on distance between bones, but I am not sure if that is a reliable approach.