diff --git a/src/chronos/base.py b/src/chronos/base.py
index daecf4c..0f4b516 100644
--- a/src/chronos/base.py
+++ b/src/chronos/base.py
@@ -165,10 +165,11 @@ class BaseChronosPipeline(metaclass=PipelineRegistry):
         quantile_levels
             Quantile levels to compute
         validate_inputs
-            When True, the dataframe(s) will be validated before prediction, ensuring that timestamps have a
-            regular frequency, and item IDs match between past and future data. Setting to False disables these checks.
+            [ADVANCED] When False, skips validation. You must ensure: (1) df and future_df (if provided)
+            are sorted by (id_column, timestamp_column); (2) future_df (if provided) contains exactly
+            prediction_length rows per item. Defaults to True.
         freq
-            Frequency string for timestamp generation (e.g., "h", "D", "W"). Can only be used when
+            Frequency string for timestamp generation (e.g., "H", "D", "W"). Can only be used when
             validate_inputs=False. When provided, skips frequency inference from the data.
         **predict_kwargs
             Additional arguments passed to predict_quantiles
diff --git a/src/chronos/chronos2/pipeline.py b/src/chronos/chronos2/pipeline.py
index f066bee..f16570a 100644
--- a/src/chronos/chronos2/pipeline.py
+++ b/src/chronos/chronos2/pipeline.py
@@ -865,8 +865,9 @@ class Chronos2Pipeline(BaseChronosPipeline):
             For optimal results, consider using a batch size around 100 (as used in the Chronos-2 technical report).
             - Cross-learning is most helpful when individual time series have limited historical context, as the model can leverage patterns from related series in the batch.
         validate_inputs
-            When True, the dataframe(s) will be validated before prediction, ensuring that timestamps have a
-            regular frequency, and item IDs match between past and future data. Setting to False disables these checks.
+            [ADVANCED] When False, skips validation. You must ensure: (1) df and future_df (if provided)
+            are sorted by (id_column, timestamp_column); (2) future_df (if provided) contains exactly
+            prediction_length rows per item. Defaults to True.
         freq
             Frequency string for timestamp generation (e.g., "h", "D", "W"). Can only be used when
             validate_inputs=False. When provided, skips frequency inference from the data.
diff --git a/src/chronos/df_utils.py b/src/chronos/df_utils.py
index 00731b2..bc09f37 100644
--- a/src/chronos/df_utils.py
+++ b/src/chronos/df_utils.py
@@ -231,10 +231,12 @@ def convert_df_input_to_list_of_dicts_input(
     timestamp_column
         Name of column containing timestamps
     freq
-        Frequency string for timestamp generation (e.g., "h", "D", "W"). Can only be used
+        Frequency string for timestamp generation (e.g., "H", "D", "W"). Can only be used
         when validate_inputs=False. When provided, skips frequency inference from the data.
     validate_inputs
-        When True, the dataframe(s) will be validated before conversion
+        [ADVANCED] When False, skips validation. You must ensure: (1) df and future_df (if provided)
+        are sorted by (id_column, timestamp_column); (2) future_df (if provided) contains exactly
+        prediction_length rows per item. Defaults to True.
 
     Returns
     -------