And then people go to empirical data and apply the Great Smoothing: By dropping ML/DL methods on the data (also including results from discontinuous behavior),

And then people go to empirical data and apply the Great Smoothing: By dropping ML/DL methods on the data (also including results from discontinuous behavior), continuity is often implicitly assumed.