diff --git a/doc/metadata_routing.rst b/doc/metadata_routing.rst
index 0a73ab803271b..b7f95f3d608d7 100644
--- a/doc/metadata_routing.rst
+++ b/doc/metadata_routing.rst
@@ -248,7 +248,8 @@ should be passed to the estimator's scorer or not::
     [sample_weight] are passed but are not explicitly set as requested or not
     requested for LogisticRegression.score, which is used within GridSearchCV.fit.
     Call `LogisticRegression.set_score_request({metadata}=True/False)` for each metadata
-    you want to request/ignore.
+    you want to request/ignore. See the Metadata Routing User guide
+    <https://scikit-learn.org/stable/metadata_routing.html> for more information.
 
 The issue can be fixed by explicitly setting the request value::
 
diff --git a/examples/model_selection/plot_cost_sensitive_learning.py b/examples/model_selection/plot_cost_sensitive_learning.py
index c4dbb64535d69..9845d27661374 100644
--- a/examples/model_selection/plot_cost_sensitive_learning.py
+++ b/examples/model_selection/plot_cost_sensitive_learning.py
@@ -689,3 +689,6 @@ def business_metric(y_true, y_pred, amount):
 # historical data (offline evaluation) should ideally be confirmed by A/B testing
 # on live data (online evaluation). Note however that A/B testing models is
 # beyond the scope of the scikit-learn library itself.
+
+# At the end, we disable the configuration flag for metadata routing::
+sklearn.set_config(enable_metadata_routing=False)
diff --git a/examples/release_highlights/plot_release_highlights_1_6_0.py b/examples/release_highlights/plot_release_highlights_1_6_0.py
index 5a1214fc31b85..7e842659f018a 100644
--- a/examples/release_highlights/plot_release_highlights_1_6_0.py
+++ b/examples/release_highlights/plot_release_highlights_1_6_0.py
@@ -69,20 +69,20 @@
 # a validation set. We can now have a pipeline which will transform the validation set
 # and pass it to the estimator::
 #
-#     sklearn.set_config(enable_metadata_routing=True)
-#     est_gs = GridSearchCV(
-#         Pipeline(
-#             (
-#                 StandardScaler(),
-#                 EstimatorWithValidationSet(...).set_fit_request(X_val=True, y_val=True),
+#     with sklearn.config_context(enable_metadata_routing=True):
+#         est_gs = GridSearchCV(
+#             Pipeline(
+#                 (
+#                     StandardScaler(),
+#                     EstimatorWithValidationSet(...).set_fit_request(X_val=True, y_val=True),
+#                 ),
+#                 # telling pipeline to transform these inputs up to the step which is
+#                 # requesting them.
+#                 transform_input=["X_val"],
 #             ),
-#             # telling pipeline to transform these inputs up to the step which is
-#             # requesting them.
-#             transform_input=["X_val"],
-#         ),
-#         param_grid={"estimatorwithvalidationset__param_to_optimize": list(range(5))},
-#         cv=5,
-#     ).fit(X, y, X_val=X_val, y_val=y_val)
+#             param_grid={"estimatorwithvalidationset__param_to_optimize": list(range(5))},
+#             cv=5,
+#         ).fit(X, y, X_val=X_val, y_val=y_val)
 #
 # In the above code, the key parts are the call to `set_fit_request` to specify that
 # `X_val` and `y_val` are required by the `EstimatorWithValidationSet.fit` method, and
diff --git a/sklearn/utils/_metadata_requests.py b/sklearn/utils/_metadata_requests.py
index ebfbc41c0eab8..d7d77a74c6fa8 100644
--- a/sklearn/utils/_metadata_requests.py
+++ b/sklearn/utils/_metadata_requests.py
@@ -458,7 +458,10 @@ def _route_params(self, params, parent, caller):
                 f" {self.owner}.{self.method}, which is used within"
                 f" {parent}.{caller}. Call `{self.owner}"
                 + set_requests_on
-                + "` for each metadata you want to request/ignore."
+                + "` for each metadata you want to request/ignore. See the"
+                " Metadata Routing User guide"
+                " <https://scikit-learn.org/stable/metadata_routing.html> for more"
+                " information."
             )
             raise UnsetMetadataPassedError(
                 message=message,
@@ -1384,7 +1387,7 @@ def __init_subclass__(cls, **kwargs):
 
         for method in SIMPLE_METHODS:
             mmr = getattr(requests, method)
-            # set ``set_{method}_request``` methods
+            # set ``set_{method}_request`` methods
             if not len(mmr.requests):
                 continue
             setattr(
diff --git a/sklearn/utils/multiclass.py b/sklearn/utils/multiclass.py
index 5df206259c5d1..6c089069387be 100644
--- a/sklearn/utils/multiclass.py
+++ b/sklearn/utils/multiclass.py
@@ -137,7 +137,7 @@ def is_multilabel(y):
     Returns
     -------
     out : bool
-        Return ``True``, if ``y`` is in a multilabel format, else ```False``.
+        Return ``True``, if ``y`` is in a multilabel format, else ``False``.
 
     Examples
     --------