Just joined AudioCircle, so my reply is late...apologies.
My impressions counter the conventional wisdom here, that XLR connectors either do not sound different or that the differences are so small as to be musically insignificant. I was also somewhat leery of various claims, but for a couple of reasons decided to investigate this for my own benefit. Those reasons include the differences that are (to me, at least) clearly audible among different RCA connectors -- the Eichmanns, the WBTs, the Xhadow and Vampires to name some of the best I've heard -- surely such differences might also be possible among XLRs; and that I use both RCA and XLR and would hope that XLR connector technology hasn't been dormant while advances have been made with RCAs.
One can discount opinions of comparisons in part as being placebo effect, but you have to be pretty cynical to assume that's all there is to such opinions. In any case, I decide to try a few of these alternative connectors (both RCA and XLR) myself, using the same wire for all cables, and even trying a few DCT'd (deep cyrogenic treated) variants of one connector to see if that made any difference. The same solder was used in all soldered cases (though solderless connections were also tried where possible). Differences were audible, but in a sense they are insignificant or, better put, small in magnitude -- they can be swamped by other things, this is best tried on a well-tuned setup. I've been working at well-tuning mine for a long time

On my system these differences are audible, but subtle. The sort of differences I hear is increased low level detail of the sort that one hears in live performance -- subtle onset transient of bowing strings or the column of air being excited in a wind instrument, or more of the inflection and vocal detail on recordings of voice. With the right source material, the music is more 'alive' with the 'right' connectors. Of the XLRs I tried, the Neutriks (which I've also used for many years) were dead last. Better were the Rhodium plated Furutech and the DCT'd Gold plated Vampire (which I preferred over the Furutech for what it's worth), and best were DCT'd Xhadow XLRs (solderless; the Take Five Audio DCT is better than an alternative source's IMO -- more nuanced with low level signal). While the differences might be judged by someone else to be so small as to be unimportant, I personally find this improved low level resolution to capture more of the beauty of timbre and to be more 'involving'. And, btw, the solderless connection does sound better, improving this revelation of low level information.
These are my conclusions so far in the proverbial nutshell. There was a bit of expense doing this but like I said, it was for selfish reasons, sometimes you have to try something to really know what's going on. After trying a few different cables and having the mindset that it's the cable and not the connector that matters, I've concluded that mindset was off the mark. Of course, YMMV.
Best regards to all...
Jeff