The difference between "data rate" and "maximum data rate" on upstream, while the SNRM remains close to 6dB, suggests that DLM has indeed intervened there - so some gains may appear in time.
The lack of a difference on downstream suggests there isn't the same problem there, and no obvious gain to come.
However, my usual rule of thumb for a line around 60Mbps is that 3dB is worth around 8Mbps downstream. If attenuation decreased by 4dB, and there were no change in the noise at the same time, you'd expect speeds to have increased around 10Mbps.
I also wonder why DLM would have intervened upstream, but not downstream. That would be a little strange, ordinarily. Are you sure G.INP isn't already active?