Is there a confusion by considering the temperature/dewpoint charts of atmospheric conditions used to illustrate the range and severity of potential carb icing which can occour within the carburettor, and applying the same chart indications to the temperature/dewpoint of the air once it is within the carburettor venturi?
If air of a given RH is heated, the RH reduces.
This means the air can hold more water as vapour and is less likely to contain water in the liquid form, consequently less likely to be able to deposit ice particles.
This means that if air was low enough RH not to form water droplets when its temperature is lowered in the venturi, then applying carb heat will only ensure the RH is even lower.
If the RH was such that the lower temperature in the carburettor could result in formation of water droplets and ice particles, adding heat will lessen the possibility.
If ice has already formed, carb heat should be sufficient to melt the ice assuming the engine power is set at a level to produce the carb heat.