diff --git a/paper.md b/paper.md
index ff368af..80d7b50 100644
--- a/paper.md
+++ b/paper.md
@@ -72,38 +72,37 @@ node sampling, and the exploration-exploitation strategies that involve building
 tree structures. It is hence time-consuming to implement and test one single algorithm.
     
  - The problem settings for the algorithms could be slightly different. As shown in Table \ref{tab: summary}, some 
-algorithms such as \texttt{HOO} and \texttt{HCT} are designed for the setting where the function evaluations can be noisy,
-while \texttt{SequOOL} is proposed for the noiseless case. Some algorithms focus on cumulative-regret optimization 
+algorithms such as \texttt{HOO} [@bubeck2011X] and \texttt{HCT} [@azar2014online] are designed for the setting where the function evaluations can be noisy,
+while \texttt{SequOOL} [@bartlett2019simple] is proposed for the noiseless case. Some algorithms focus on cumulative-regret optimization 
 whereas some only care about the last-point regret or the simple regret\footnote{ A more detailed discussion on simple
 regret and cumulative regret can be found in [@bubeck2011X]}. Therefore, experimental comparisons often focus on a small
 subset of algorithms, see e.g., [@azar2014online], [@bartlett2019simple]. The unavailability of a general package only 
 deteriorates the situation. 
 
-
 \begin{table}
     \centering
     \caption{Selected examples of $\mathcal{X}$-armed bandit algorithms implemented in our library. \textit{Cumulative}: whether the algorithm focuses on optimizing cumulative regret or simple regret. \textit{Stochastic}: whether the algorithm deals with noisy rewards. \textit{Open-sourced?}:  the code availability before the development of PyXAB.}
     \begin{tabular}{l c c c}
+        \hline
+        {$\mathcal{X}$-Armed Bandit Algorithm} 
+        & Cumulative & Stochastic & {Open-sourced?}   \\
+        \hline
+         \texttt{HOO} & yes & yes & yes (Python)  \\
+        \texttt{DOO}   & no & no & no  \\
+        \texttt{StoSOO}   & no & yes & yes (MATLAB, C)  \\
+        \texttt{HCT}   & yes & yes & no   \\
+        \texttt{POO}  & no & yes & yes (Python, R) \\
+        \texttt{GPO}   & no & yes & no  \\
+        \texttt{SequOOL}   & no & no &no \\
+        \texttt{StroquOOL}    & no & yes &no \\
+        \texttt{VROOM}   & no & no &no \\
+        \texttt{VHCT}   & yes & yes &no\\
+       \hline
     \end{tabular}
+        \vspace{-10pt}
     \label{tab: summary}
-
-| $\mathcal{X}$-Armed Bandit Algorithm       | Cumulative | Stochastic | Open-sourced?    |
-|--------------------------------------------|:----------:|:----------:|------------------|
-| \texttt{HOO} [@bubeck2011X]                |    yes     |    yes     | yes (Python)     |
-| \texttt{DOO} [@Munos2011Optimistic]        |     no     |     no     | no               |
-| \texttt{StoSOO} [@Valko13Stochastic]       |     no     |    yes     | yes (MATLAB)     |
-| \texttt{HCT} [@azar2014online]             |    yes     |    yes     | no               |
-| \texttt{POO} [@Grill2015Blackbox]          |     no     |    yes     | yes  (Python, R) |
-| \texttt{GPO} [@shang2019general]           |     no     |    yes     | no               |
-| \texttt{SequOOL}  [@bartlett2019simple]    |     no     |     no     | no               |
-| \texttt{StroquOOL}  [@bartlett2019simple]  |     no     |    yes     | no               |
-| \texttt{VROOM}  [@ammar20derivative]       |     no     |     no     | no               |
-| \texttt{VHCT}  [@li2021optimumstatistical] |    yes     |    yes     | no               |
-
 \end{table}
 
-
-
 To remove the barriers for future research in this area, we have developed PyXAB, a Python library of the existing 
 popular $\mathcal{X}$-armed bandit algorithms. To the best of our knowledge, this is the first comprehensive library for
 $\mathcal{X}$-armed bandit, with clear documentations and user-friendly API references.