diff --git a/lms/templates/courseware/instructor_dashboard.html b/lms/templates/courseware/instructor_dashboard.html
index aa3c33e43a..561c6b9507 100644
--- a/lms/templates/courseware/instructor_dashboard.html
+++ b/lms/templates/courseware/instructor_dashboard.html
@@ -198,7 +198,8 @@ function goto( mode)
     <H2>Student-specific grade inspection and adjustment</h2>
     <p>edX email address or their username: </p>
     <p><input type="text" name="unique_student_identifier">  <input type="submit" name="action" value="Get link to student's progress page"></p>
-    <p>and, if you want to reset the number of attempts for a problem, the urlname of that problem</p>
+    <p>and, if you want to reset the number of attempts for a problem, the urlname of that problem 
+    (e.g. if the location is <tt>i4x://university/course/problem/problemname</tt>, then the urlname is <tt>problemname</tt>).</p>
     <p> <input type="text" name="problem_to_reset" size="60"> <input type="submit" name="action" value="Reset student's attempts">  </p>
 
     %if instructor_access:
diff --git a/lms/templates/instructor/staff_grading.html b/lms/templates/instructor/staff_grading.html
index 1c5f7364ad..0a28a2b026 100644
--- a/lms/templates/instructor/staff_grading.html
+++ b/lms/templates/instructor/staff_grading.html
@@ -29,7 +29,7 @@
 <section class="problem-list-container">
 <h2>Instructions</h2>
 <div class="instructions">
-<p>This is the list of problems that current need to be graded in order to train the machine learning models. Each problem needs to be trained separately, and we have indicated the number of student submissions that need to be graded in order for a model to be generated. You can grade more than the minimum required number of submissions--this will improve the accuracy of machine learning, though with diminishing returns. You can see the current accuracy of machine learning while grading.</p>
+<p>This is the list of problems that currently need to be graded in order to train the machine learning models. Each problem needs to be trained separately, and we have indicated the number of student submissions that need to be graded in order for a model to be generated. You can grade more than the minimum required number of submissions--this will improve the accuracy of machine learning, though with diminishing returns. You can see the current accuracy of machine learning while grading.</p>
 </div>
 
 <h2>Problem List</h2>