TLDR: unlikely without adviser, but expository papers are a possibility
I have done/did undergraduate research in Physics, Chemistry, and Mathematics. And one certain thing I can tell you is that without research advisor, the chances of doing research is slim to none. First of all, you need to understand that in a good paper, while an undergraduate might contribute a lot in terms of grunt work, he will have no idea about how to state the sub-sub-sub problem in some obscure sub-sub-sub field and how the problem came about in the first place. For example, when I did research in chemistry, I was told everything and all I did was lock myself in the lab and take trials after trials, after which it was writing scripts in python to do statistical analysis. You ride the coattail of your adviser. That's the simple truth of it.
If it's theoretical research (see physics or pure math), even with an adviser you won't do much except maybe follow the steps that are more-or-less outlined for you. For a typical undergraduate, that is. [For reference, I remember reading Peter Clark's (UGA tenured mathematician, active user on stackexchange) answer/comment where he said how recommendations from REU advisers don't carry weight because 99% of the time students really don't make real contributions in 8 weeks where they have to learn the material to begin with]. However, they are good experiences nevertheless.
Another thing one could do without adviser is: write expository papers in advanced topics. Certainly, having an adviser is better. That'll usually require lots of reading at graduate level and could be good. [For reference: many papers of UChicago REU are often expository papers]