Papers increasingly come with a supplementary material. This is in principle good as providing all the gory details, except that often this is where the authors flush out what didn't fit in the allotted space of the main text, so supplementary of PRL, for instances, are particularly tough (as often required-reading)
When you upload your prose to arXiv, you typically want the supplementary to be there as well. That's not easy to do.
For some reason, LaTeX (revTeX, actually) makes it complicated to merge two documents. The \maketitle, in particular, has a relax option, which means once it's been invoked once, you can forget about it. Now, the supplementary should come with its own title, list of authors, affiliations, etc. It is, technically, a separate (supplementary) paper.
Solutions proposed online are not particularly pretty [1][2]. You can't really hack the revtex.cls file... as you wouldn't like to upload this as well as part of your submission, although that would technically work. Using the titling package is not really useful either because it implies so much reformatting to make it look like what it should...
The best solution I can find is to do it yourself, by hand.
E.g., that's the normal declarations for a recent paper that faced this dilemma:
\title{Negative-mass effects in spin-orbit coupled Bose-Einstein condensates}
\author{David Colas}
\email{d.colas@uq.edu.au}
\affiliation{ARC Centre of Excellence in Future Low-Energy Electronics Technologies, School of Mathematics and Physics, University of Queensland, St Lucia, Queensland 4072, Australia}
\author{Fabrice P. Laussy}
\affiliation{Faculty of Science and Engineering, University of Wolverhampton, Wulfruna St, Wolverhampton WV1 1LY, United Kingdom}
\affiliation{Russian Quantum Center, Novaya 100, 143025 Skolkovo, Moscow Region, Russia}
\author{Matthew J. Davis}
\affiliation{ARC Centre of Excellence in Future Low-Energy Electronics Technologies, School of Mathematics and Physics, University of Queensland, St Lucia, Queensland 4072, Australia}
\begin{abstract}
Negative effective masses can be realised by engineering the
dispersion relation in a variety of quantum systems. A recent
experiment with spin-orbit coupled Bose-Einstein condensates has
shown that a negative effective mass can halt the free expansion of
the condensate and produce fringes in the density [M.~Khamehchi
\textit{et al.}, Phys. Rev. Lett. \textbf{118}, 155301 (2017)].
Here, we show that the observed fringes are due to the negativity of
only one of the two effective mass parameters that characterise the
dispersion, which leads to previously predicted self-interference
of the wave packet. We show how configurations are nevertheless
accessible to spin-orbit coupled Bose-Einstein condensates where
both mass parameters controlling the propagation and diffusion of
the condensate are negative, resulting in the novel phenomenon of
counter-propagating self-interfering packets.
\end{abstract}
\pacs{} \date{\today} \maketitle
and at the end of the text, before \end{document}, this is the bit to start putting the Supplementary Material:
%%%%%%%%%%%%%%%% SUPPLEMENTARY
\pagebreak
\onecolumngrid
\begin{center}
\textbf{\large Negative-mass effects in spin-orbit coupled Bose-Einstein condensates\\Supplementary Material}\\[.2cm]
David Colas,$^{1,*}$ Fabrice P. Laussy,$^{2,3}$ and Matthew J. Davis$^1$\\[.1cm]
{\itshape ${}^1$ARC Centre of Excellence in Future Low-Energy Electronics Technologies,\\ School of Mathematics and Physics, University of Queensland, St Lucia, Queensland 4072, Australia\\
${}^2$Faculty of Science and Engineering, University of Wolverhampton,\\Wulfruna St, Wolverhampton WV1 1LY, United Kingdom\\
${}^3$Russian Quantum Center, Novaya 100, 143025 Skolkovo, Moscow Region, Russia\\}
${}^*$Electronic address: d.colas@uq.edu.au\\
(Dated: \today)\\[1cm]
\end{center}
\twocolumngrid
\setcounter{equation}{0}
\setcounter{figure}{0}
\setcounter{table}{0}
\setcounter{page}{1}
\renewcommand{\theequation}{S\arabic{equation}}
\renewcommand{\thefigure}{S\arabic{figure}}
\renewcommand{\bibnumfmt}[1]{[S#1]}
\renewcommand{\citenumfont}[1]{S#1}
The \twocolumngrid can be removed if you want to keep the supplementary single-columned as it may be useful to strongly separate it visually from the main text. After that, equations are reset to zero and to avoid confusion, are labelled differently, so that one does not mix material from the text and the supplementary. Here, this will have number equations (S1), (S2), etc., and references [S1], etc.