My partner is a very bright theoretical CS student. Recently she proved a pretty significant result that she is aiming to submit for SODA. In addition, the paper was done completely without her advisor (who is a very well established tenured professor in the field).
She has shared with me that she is very worried about publishing the paper alone, in case the paper has a mistake. She wants to publish it with her advisor, so that he spends some time reading, verifying the result, and rewriting some rough patches in the paper. I've told her that publishing as a single author PhD student is very impressive in this field, and her advisor has encouraged her to publish it alone, if she so desires.
My question is, how big of a problem is it if she publishes it and there turns our to be a mistake? Is it very deterimental to her reputation? I've even tried to read the paper to help her, and I believe to the best of my knowledge that the results are correct, but then again, I'm not from the community so I cannot provide her with the assurance her advisor would otherwise give her that the results make sense and are correct. It also took me more than 3 weeks to verify the results (or roughly $30$ hours), so it is a big investment of time that her advisor would have to put in.
She obviously believes her results are correct, but is lacking on the confidence to let the paper out. What would be the sane thing to do here? Submit anonymously to SODA and see the reviews instead of posting on arxiv?