The new release includes many bug fixes--they were legion--but also some improved functionality.
Direct classification routines will now return the joint probabilities in addition to conditional probabilities. Since you can use either for classification, the joint probabilities provide more information.
It is now possible to search for a class border at some point other than R=0. This is useful if the classes differ greatly in size or if the relative number of samples do not reflect the actual class sizes.
You can use other metrics than Cartesian. There are some issues with that.
We now have n-fold cross validation!--for the classification routines, at least. It still needs to be added for the interpolation/non-linear regression routines.
We also have clustering. This will be detailed in a subsequent post.