diff --git a/.gitignore b/.gitignore
index 4ffab5fb4..b346a410b 100644
--- a/.gitignore
+++ b/.gitignore
@@ -15,6 +15,7 @@ __pycache__
 docs/_build/
 docs/src/generated/
 docs/src/sg_execution_times.rst
+# ignore file copied during doc build
 tools/documentation_developer/CONTRIBUTING.rst
 
 # file generated by hatch
diff --git a/CONTRIBUTING.rst b/CONTRIBUTING.rst
index 76e877f91..6616b114c 100644
--- a/CONTRIBUTING.rst
+++ b/CONTRIBUTING.rst
@@ -21,7 +21,7 @@ For testing, we recommend you install the test dependencies with pip:
 
 .. code-block:: sh
 
-   pip install hidimstat[test]
+   pip install -e '.[test]'
   
 This will install ``pytest`` and the following extensions: 
 ``pytest-cov``, ``pytest-randomly``, ``pytest-xdist``, ``pytest-html``,
@@ -40,7 +40,7 @@ To run the examples, we recommend you install the example dependencies with pip:
 
 .. code-block:: sh
 
-    pip install hidimstat[example]
+    pip install -e '.[example]'
 
 For running the examples, it's necessary to install Matplotlib >= |MatplotlibMinVersion| and seaborn >=
 |SeabornMinVersion|.
@@ -55,7 +55,7 @@ The documentation is built with Sphinx. We recommend you install the documentati
 
 .. code-block:: sh
 
-    pip install hidimstat[doc]
+    pip install -e '.[doc]'
 
 After this installation, you can build the documentation from the source using the Makefile in doc_conf:
 
@@ -63,4 +63,4 @@ After this installation, you can build the documentation from the source using t
 
     make html
 
-**For more information**, look at the `developer documentation <https://hidimstat.github.io/dev/dev/index.html>`_
\ No newline at end of file
+**For more information**, look at the `developer documentation <https://hidimstat.github.io/dev/dev/index.html>`_
diff --git a/docs/src/index.rst b/docs/src/index.rst
index 8a82a4df0..b314ae90d 100644
--- a/docs/src/index.rst
+++ b/docs/src/index.rst
@@ -172,8 +172,8 @@ For Knockoffs Inference:
    :target: https://github.com/psf/black
 
 
-References
-----------
+Bibliography
+------------
 .. bibliography:: ../tools/references.bib
 
   
diff --git a/docs/tools/conf.py b/docs/tools/conf.py
index 6475f3930..51623e771 100644
--- a/docs/tools/conf.py
+++ b/docs/tools/conf.py
@@ -96,7 +96,7 @@
 
 # The suffix(es) of source filenames.
 # You can specify multiple suffix as a list of string:
-source_suffix = [".rst", ".md"]
+source_suffix = {".rst": "restructuredtext"}
 
 # sphinxcontrib-bibtex
 bibtex_bibfiles = ["../tools/references.bib"]
diff --git a/examples/plot_2D_simulation_example.py b/examples/plot_2D_simulation_example.py
index 82b888146..20e30cb5f 100644
--- a/examples/plot_2D_simulation_example.py
+++ b/examples/plot_2D_simulation_example.py
@@ -41,10 +41,6 @@
 significantly the spatial uncertainty compared to clustered inference
 algorithms which consider only one spatial compression.
 
-References
-----------
-.. footbibliography::
-
 """
 import matplotlib.pyplot as plt
 
@@ -356,3 +352,9 @@ def plot(maps, titles):
 # conservative. In practice, Type-1 Error guarantees seem to hold
 # for a lower spatial tolerance. This is an additional benefit of clustering
 # randomization.
+
+#############################################################################
+# References
+# ----------
+# .. footbibliography::
+#
diff --git a/examples/plot_diabetes_variable_importance_example.py b/examples/plot_diabetes_variable_importance_example.py
index 0340d9a3d..fd8bd3a3a 100644
--- a/examples/plot_diabetes_variable_importance_example.py
+++ b/examples/plot_diabetes_variable_importance_example.py
@@ -36,10 +36,6 @@
 single-level case. The aim is to see if integrating the new
 statistically-controlled solution has an impact on the results.
 
-References
-----------
-.. footbibliography::
-
 """
 
 #############################################################################
@@ -255,3 +251,10 @@ def compute_pval(vim):
 ax.set_xticks(range(len(diabetes.feature_names)))
 ax.set_xticklabels(diabetes.feature_names)
 plt.show()
+
+
+#############################################################################
+# References
+# ----------
+# .. footbibliography::
+#
diff --git a/src/hidimstat/distilled_conditional_randomization_test.py b/src/hidimstat/distilled_conditional_randomization_test.py
index cd76ed971..1ce24ef0d 100644
--- a/src/hidimstat/distilled_conditional_randomization_test.py
+++ b/src/hidimstat/distilled_conditional_randomization_test.py
@@ -403,7 +403,7 @@ def fit_importance(self, X, y, cv=None):
 
         Notes
         -----
-        Also sets the importances\_ and pvalues\_ attributes on the instance.
+        Also sets the ``importances_`` and ``pvalues_`` attributes on the instance.
         See fit() and importance() for details on the underlying computations.
         """
         if cv is not None: