room topic changed to amateur reviews: paper reviews and discussions about non-peer-reviewed results (no tags)

Papers on arXiv trying to proof ambitious results are a dime a dozen. Still, sometimes I read such papers, sometimes I communicate with the authors of a paper. Sometimes people query about the validity of the results from such papers.

Feel free to discuss papers here on any level you like, ranging from analysing prior publications of its authors, the academic status of its authors, whether authors made appropriate references to existing literature, to details of the paper and the exact location where the first major mistake occurs.

The first paper I plan to review is A Message-Passing Algorithm for Graph Isomorphism, which was brought up by user Turbo. I haven't read the paper in detail yet. I see that the author implemented his algorithm and compared its performance to Nauty, which is already a good sign that reading this paper might be worth my time.