diff --git a/CHANGES/v2.4.md b/CHANGES/v2.4.md
index ef91fd3532d032d23921dd7a05fded55af276476..a0b816fa81fab4627456256ea4004e63ed3f0d04 100644
--- a/CHANGES/v2.4.md
+++ b/CHANGES/v2.4.md
@@ -233,7 +233,7 @@ For users:
 ## Version 2.4.6 ()
 
 For users:
-  - Fixed a bug in \ref COORDINATIONNUMBER where derivatives were wrong when using R_POWER > 2, thanks to @MoleOrbitalHybridAnalyst for spotting and fixing
+  - Fixed a bug in \ref COORDINATIONNUMBER where derivatives were wrong when using R_POWER > 2, thanks to `@MoleOrbitalHybridAnalyst` for spotting and fixing
   - Fixed a bug in library search, possibly affecting linked blas/lapack on OSX (see \issue{476}).
   - Fixed a bug in \ref METAD with `TARGET` and `GRID_SPARSE` (see \issue{467}).
 
diff --git a/user-doc/tutorials/performance-optimization.txt b/user-doc/tutorials/performance-optimization.txt
index c6d42929bbb7d70d7b0d01e5e634188ebd4d6922..804f9b0215d4bb02423602e58076c51a2715b95f 100644
--- a/user-doc/tutorials/performance-optimization.txt
+++ b/user-doc/tutorials/performance-optimization.txt
@@ -175,7 +175,7 @@ PRINT ARG=dp,cn FILE=COLVAR STRIDE=100
 \endplumedfile
 
 Now measure the speed of your simulation. You should see some overhead due to PLUMED.
-You can now try to use multiple-time-stepping \cite{Ferrarotti2015}.
+You can now try to use multiple-time-stepping \cite Ferrarotti2015.
 To do so add a `STRIDE` option to the \ref RESTRAINT line.
 You can also use the \ref EFFECTIVE_ENERGY_DRIFT action to print a kind of "total energy drift" due to the application
 of the PLUMED bias. In molecular dynamics simulation you usually monitor the drift in the total energy in order to choose the
@@ -214,7 +214,7 @@ Using a `STRIDE=2` instead the effective energy will be very stable
 
 
 Let's try to perform a metadynamics simulation biasing the coordination number and
-the distance between the magnesium ion and the phosphate. Parameters are similar to those used in \cite{cunha2017unraveling},
+the distance between the magnesium ion and the phosphate. Parameters are similar to those used in \cite cunha2017unraveling,
 although we use here a shorter deposition pace in order to artificially increase the computational cost.
 
 \plumedfile