Having just read the Heaters section by Dingwall in the RCA 1962 Electron Tube Design book from the following link, I reckon that reference would just about be the most informative assessment to work from.
Electron Tube Design : Radio Corporation of America : Free Download & Streaming : Internet Archive
Dingwall discusses how hugely varied the level of leakage can be, even from in the same batch, and how it is not typically asymmetric around zero volts in just one particular manner. The contribution to leakage from electron flow is also shown as being just one of a mix of many mechanisms that can contribute to leakage current.
Maybe its my misinterpretation of your sentence, but I think you're saying that DC bias causes a DC current in only one direction, which is fine, but the inference is then that that dc current somehow reduces or stops AC leakage currents from flowing and causing hum. Maybe it's clearer to say that a particular DC voltage bias will cause a certain heater-cathode resistance value, and that resistance value then determines the level of AC hum current flowing. The point being that a higher heater-cathode resistance will lower the AC hum current flowing, and the typical characteristic is that heater cathode resistance increases as the DC voltage bias increases (either in a positive or negative sense, as it is not generic to just one polarity), and by biasing beyond about +/- 20VDC the combined DC plus heater AC voltage doesn't get smaller than about 10V during the AC cycle, and so the effective resistance seen by the AC voltage is adequately higher.
The take away message I get is that there can be a huge variation in the heater-cathode resistance between even the same type of tube, but in general the resistance is so high as to be of no consequence from a hum perspective, as other hum contributors (capacitive coupling from heater to grid, or from heater to cathode) are normally more dominant.
Electron Tube Design : Radio Corporation of America : Free Download & Streaming : Internet Archive
Dingwall discusses how hugely varied the level of leakage can be, even from in the same batch, and how it is not typically asymmetric around zero volts in just one particular manner. The contribution to leakage from electron flow is also shown as being just one of a mix of many mechanisms that can contribute to leakage current.
Maybe its my misinterpretation of your sentence, but I think you're saying that DC bias causes a DC current in only one direction, which is fine, but the inference is then that that dc current somehow reduces or stops AC leakage currents from flowing and causing hum. Maybe it's clearer to say that a particular DC voltage bias will cause a certain heater-cathode resistance value, and that resistance value then determines the level of AC hum current flowing. The point being that a higher heater-cathode resistance will lower the AC hum current flowing, and the typical characteristic is that heater cathode resistance increases as the DC voltage bias increases (either in a positive or negative sense, as it is not generic to just one polarity), and by biasing beyond about +/- 20VDC the combined DC plus heater AC voltage doesn't get smaller than about 10V during the AC cycle, and so the effective resistance seen by the AC voltage is adequately higher.
The take away message I get is that there can be a huge variation in the heater-cathode resistance between even the same type of tube, but in general the resistance is so high as to be of no consequence from a hum perspective, as other hum contributors (capacitive coupling from heater to grid, or from heater to cathode) are normally more dominant.