diff --git a/CHANGELOG.md b/CHANGELOG.md
index 1e54044..e62a1d3 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -25,6 +25,7 @@ and this project adheres to [Semantic Versioning][].
 -   Fix neighbors connectivities in test to use new scanpy fn {pr}`170`.
 -   Fix Kmeans test {pr}`172`.
 -   Fix deprecation and future warnings {pr}`171`.
+-   Fix lisi return type and docstring {pr}`182`.
 
 ## 0.5.1 (2024-02-23)
 
diff --git a/src/scib_metrics/metrics/_lisi.py b/src/scib_metrics/metrics/_lisi.py
index 17c2168..9c088d4 100644
--- a/src/scib_metrics/metrics/_lisi.py
+++ b/src/scib_metrics/metrics/_lisi.py
@@ -39,7 +39,7 @@ def lisi_knn(X: NeighborsResults, labels: np.ndarray, perplexity: float = None)
     return 1 / simpson
 
 
-def ilisi_knn(X: NeighborsResults, batches: np.ndarray, perplexity: float = None, scale: bool = True) -> np.ndarray:
+def ilisi_knn(X: NeighborsResults, batches: np.ndarray, perplexity: float = None, scale: bool = True) -> float:
     """Compute the integration local inverse simpson index (iLISI) for each cell :cite:p:`korsunsky2019harmony`.
 
     Returns a scaled version of the iLISI score for each cell, by default :cite:p:`luecken2022benchmarking`.
@@ -60,7 +60,7 @@ def ilisi_knn(X: NeighborsResults, batches: np.ndarray, perplexity: float = None
     Returns
     -------
     ilisi
-        Array of shape (n_cells,) with the iLISI score for each cell.
+        iLISI score.
     """
     batches = np.asarray(pd.Categorical(batches).codes)
     lisi = lisi_knn(X, batches, perplexity=perplexity)
@@ -71,7 +71,7 @@ def ilisi_knn(X: NeighborsResults, batches: np.ndarray, perplexity: float = None
     return ilisi
 
 
-def clisi_knn(X: NeighborsResults, labels: np.ndarray, perplexity: float = None, scale: bool = True) -> np.ndarray:
+def clisi_knn(X: NeighborsResults, labels: np.ndarray, perplexity: float = None, scale: bool = True) -> float:
     """Compute the cell-type local inverse simpson index (cLISI) for each cell :cite:p:`korsunsky2019harmony`.
 
     Returns a scaled version of the cLISI score for each cell, by default :cite:p:`luecken2022benchmarking`.
@@ -92,7 +92,7 @@ def clisi_knn(X: NeighborsResults, labels: np.ndarray, perplexity: float = None,
     Returns
     -------
     clisi
-        Array of shape (n_cells,) with the cLISI score for each cell.
+        cLISI score.
     """
     labels = np.asarray(pd.Categorical(labels).codes)
     lisi = lisi_knn(X, labels, perplexity=perplexity)