diff --git a/paper/main.md b/paper/main.md
index 9077687..2ec48e1 100644
--- a/paper/main.md
+++ b/paper/main.md
@@ -932,7 +932,7 @@ We see three main prospects for future work:
   Additionally some expandable per-plugin store for the solvers would be handy, for example to not recompute type class instances on every invocation.
   Somewhat more challenging, one can imagine a caching system for constraints, allowing to avoid solving the same constraint twice.
   This might be beneficial for reduction, since as of now we do a lot of redundant computations.
-  However, the memory usage might be prohibitive.
+  However, memory usage might be prohibitive.
   Finally, we would also like to explore possibilities for concurrent solving, similar to the future plans of @allaisTypOSOperatingSystem2022a with LVars for metavariables [@kuperLatticebasedDataStructures2015].
 
 ::: {#refs}
diff --git a/paper/main.tex b/paper/main.tex
index be85d25..814cbef 100644
--- a/paper/main.tex
+++ b/paper/main.tex
@@ -1563,7 +1563,7 @@ \section{Future work}\label{future-work}}
   Somewhat more challenging, one can imagine a caching system for
   constraints, allowing to avoid solving the same constraint twice. This
   might be beneficial for reduction, since as of now we do a lot of
-  redundant computations. However, the memory usage might be
+  redundant computations. However, memory usage might be
   prohibitive. Finally, we would also like to explore possibilities for
   concurrent solving, similar to the future plans of
   \citet{allaisTypOSOperatingSystem2022a} with LVars for metavariables