The UKC / etc type displays aren't like the wattmeters--they do not have current-measuring shunts in them, so they don't see any of the voltage generated in them, whether positive or negative.
However, if your controller regenerates voltages above the limit of the LVPS in the display (the circuit that takes battery voltage and converts it to 5v, etc for all the display electronics), it could damage or destroy the display (or anything else connected to the battery voltage that can't handle this). It only takes one momentary spike above this limit to destroy the electronics, and you probably can't see this spike on your voltmeter, especially if it's digital, since they all average the incoming voltages over time, *and* have a sampling rate that can miss some short inputs, *and* have a display refresh rate that changes what you see even less often than any of those.
So you can look at the specs for the display (and controller, and anything else running off battery voltage) to see what they list as a maximum voltage, then try to measure the actual peak (not just displayed!) voltage generated by your regen braking under worst-case (highest regen amount at fullest battery charge voltage). (some good multimeters have a Vmax / peak hold function good enough for this). If the generated voltage is higher than the limit, or even right at it, it is probably what is causing the damage.
If the controller has a setting to limit regen voltage, you can try that by setting it as far below the lowest voltage spec limit as necessary to prevent damage. Many controllers simply prevent regen from occuring when above a regen voltage limit, rather than actually limiting the voltage, so keep that in mind when using regen braking--there may be times it simply doesn't engage.
If it's just a very short spike causing the damage, you could try a choke on the main battery input to the display. (choke = inductor in series).