\documentclass[varwidth=598pt]{standalone}

\usepackage{booktabs}      % professional-quality tables
\usepackage{multirow}      % multi-row cells in tables
\usepackage{colortbl}      % color for tables (\cellcolor, \rowcolor)
\usepackage[table]{xcolor} % enhanced colors for tables
\usepackage{array}         % more flexible column formats (often used)
\usepackage{tabularx}      % for tables with auto-stretch columns (if used)
\usepackage{graphicx}      % for images in tables or figure/table floats
\usepackage{amssymb}       % math symbols
\usepackage{amsmath}       % math environments
\usepackage{soul}          % highlighting (used for colored text/cells)
\usepackage[normalem]{ulem} % underlining, strikethroughs
\usepackage[T1]{fontenc}   % font encoding
\usepackage[utf8]{inputenc}% input encoding (legacy, fine for pdflatex)
\usepackage{microtype}     % better text appearance
\usepackage{hyperref}      % hyperlinks
\usepackage{textcomp}      % for extra symbols
\usepackage{enumitem}      % for compact lists (if used in table notes)
\usepackage{adjustbox}
\usepackage{tabu}
\usepackage{pifont}        % http://ctan.org/pkg/pifont
\usepackage{bbding}        % \XSolidBrush
\usepackage{makecell}

\begingroup
\makeatletter
\renewcommand{\fnum@table}{}%
\long\def\@makecaption#1#2{%
  \vskip\abovecaptionskip
  \centering #2\par
  \vskip\belowcaptionskip
}
\makeatother

\newcommand{\appendixhead}{\begin{center}\textbf{{\large Appendix}}\end{center}
}
\newcolumntype{R}[2]{    >{\adjustbox{angle=#1,lap=\width-(#2)}\bgroup}    l    <{\egroup}}
\newcommand*\rot{\multicolumn{1}{R{45}{1.4em}}}% no optional argument here, please!optional argument here, please!
\begin{document}
\begin{table}
\centering
\begin{adjustbox}{max width=\textwidth}
\begin{tabular}{lcccccccccc}
\hline
\multirow{2}{*}{Method}  & \multicolumn{2}{c}{CIFAR100} & \multicolumn{2}{c}{ImageNet100} & \multicolumn{2}{c}{ImageNet-R} & \multicolumn{2}{c}{CUB200} & \multicolumn{2}{c}{VTAB}\\ 
&  \textbf{Avg} $\uparrow$ & \textbf{Last} $\uparrow$ & \textbf{Avg} $\uparrow$ & \textbf{Last} $\uparrow$ & \textbf{Avg} $\uparrow$ & \textbf{Last} $\uparrow$  & \textbf{Avg} $\uparrow$ & \textbf{Last} $\uparrow$ & \textbf{Avg} $\uparrow$ & \textbf{Last} $\uparrow$ \\
\midrule
Single-task JOINT  & \multicolumn{2}{c}{80.28} & \multicolumn{2}{c}{81.08} & \multicolumn{2}{c}{80.92}  &  \multicolumn{2}{c}{75.4} &  \multicolumn{2}{c}{89.29} \\
Task-specific JOINT & \multicolumn{2}{c}{82.9} & \multicolumn{2}{c}{83.55} & \multicolumn{2}{c}{83.07}  &  \multicolumn{2}{c}{85.72} &  \multicolumn{2}{c}{94.6} \\
\midrule
iCaRL rebuffi2017icarl & 72.93  & 57.6 & 68.62& 59.5 & 66.34 & 43.71  & 82.39& 75.1 & 53.38 & 41.6  \\
\midrule
L2P wang2022learning   & 78.92& 70.04 & - & - & 77.07 & 69.33  & 76.98 & 68.47&  - & -\\
DualPrompt wang2022dualprompt  & 82.11 & 74.31 & - & - & 82.73 & 76.41  &  82.37& 76.29& - & - \\
CODA-P smith2023coda & 85.19 & 76.4 & 85.93 & 79.02 & 82.06 & 79.5 & 84.77 & 80.39 & 87.5 & 81.2 \\
PROOF zhou2023learning  & 84.84 & 76.55 & - & - & 84.89 & 79.7  &  83.98& 79.35& - & -\\
\midrule
Continual-CLIP Thengane2022CLIPMI &  78.65 & 68.26 & 83.99 & 74.2 & 84.43 & 76.94  & 67.0 & 54.8 & 68.5& 60.97   \\
CoOp zhou2022learning & 81.17 & 70.58 & 79.14 & 64.9 & 84.7 & 78.66  & 76.62 & 68.53&  87.06 & 81.25 \\
MaPLe khattak2023maple & 82.74 & 74.52 & 79.23 & 64.06& 85.28 & 79.71 & 73.38 & 64.43 & 83.91 & 81.81 \\
AttriCLIP wang2023attriclip & 79.31 & 68.45 & 82.29 & 70.76 & 83.09 & 76.53  & 65.26  & 52.12 & 71.84 & 64.09 \\
CLIP-Adapter gao2021clip  & 78.75 & 68.32 & 84.13 & 73.96 & 84.49 & 78.1 & 67.41 & 54.49 & 68.23  & 61.02 \\
VPT derakhshani2023variational & 73.4 & 59.33 & 80.51 & 61.09 & 81.66 & 74.0 & 69.14 & 60.03 & 67.2 & 77.26 \\
\midrule

Ours w/o VI & 84.36 & 76.8 & 86.11 & 76.48 & \textbf{86.69} & \textbf{80.83} & 72.21 & 61.87 & 90.74 & 88.64   \\  

\midrule
Ours  & \textbf{86.13} & \textcolor{blue}{78.21} & \textbf{87.76}  & \textcolor{blue}{79.16} & 85.77 & 79.98 & \textcolor{blue}{86.93} & \textcolor{blue}{81.64}  & \textcolor{blue}{91.37} & \textcolor{blue}{89.67} \\
CoOp + Ours  & 85.71  & 77.4 & 86.8& 78.18& 85.32 & 79.52  & \textbf{86.99} & \textbf{81.95} & \textbf{92.51} & \textbf{91.28}  \\

MaPLe + Ours & \textcolor{blue}{86.06} & \textbf{78.48} & \textcolor{blue}{87.47} & 79.02 & 86.25 & 80.56 & 81.53 & 74.24 & 90.97 & 88.83 \\

AttriCLIP + Ours  & 78.06 & 67.59 &  87.37 & \textbf{79.3} & \textcolor{blue}{86.35} & \textcolor{blue}{80.6}  & 83.71 & 79.01 & 74.84 & 71.12 \\


\bottomrule
\end{tabular}
\end{adjustbox}
\caption{Table 1 : \textbf{Performance comparison} of different methods averaged over three runs. Best scores are in \textbf{bold}. The second-best scores are in \textcolor{blue}{blue}. The results for L2P, DualPrompt, and PROOF are taken from zhou2023learning. See App. Table tab:stddev for statistical significance of these results using std. dev. scores.}
\end{table}
\end{document}