IANAScientist ...
No. The purpose of a conference [1] is to have researchers present their results. The timing guidelines for a typical conference talk would look like: 15 minutes for the talk, 5 minutes for questions and comments, and 5 minutes for setting up the next talk. As you can imagine, this doesn't give much time for discussions; dissecting a paper is not part of the agenda. The questions and comments raised in the 5-minute period after the talk usually are along the lines of "Do you think your approach will work for this other problem?", rather than "Can you explain how this step in the argument works?". If you have a question of the latter kind, then you are supposed to discuss it with the author offline, so that the conference schedule is not affected. The conference does not provide you with a forum to discuss papers at any depth.
[1] All of this pertains to Theoretical Computer Science conferences; I suspect conferences in other fields are similar, but I don't know for sure.
So basically there is a lack of pre-conference peer review.
All of this is supposed to happen before the paper is accepted for publication. The proceedings of top conferences in most fields are, in my opinion, more important than the actual conference itself.