Can you see that the contact tip is melted on the end, or does it look like the wire is arcing and melting the inside of the contact tip?
Is this a new welding application or did this problem just crop up, and it used to work fine?
Did you say what type of gas you are using?
I am wondering if possibly your arc length is excessive, causing the contact tip to overheat or the wire to begin to weld to the tip. Is it possible the arc is momentarily too long, just when you strike the arc, or when you shut off the weld? Some welding systems have arc start and wire burn back settings that affect arc length at the weld start and stop.
Is the arc length excessive during the entire weld? If the sound of the spray is totally smooth, hissing, without any crackle, it may be too long of an arc, ie, voltage setting too high or tip to work distance too long. If the wire is 0.045, then I think the wire speed of 250 to 300 ipm may be low for a voltage of 25.4 V. I could be mistaken for your particular setup and power supply, but I have run a pulsed power supply in an automated machine with 0.045 E70S-6, 92 Argon/ 8 CO2, 400 ipm, tip to work distance of 5/8", and achieved a slightly crackling spray transfer, with a very tight arc length, and a voltage reading of around 25 V.
Hope this gives you some ideas.