diff --git a/tensorflow_datasets/core/utils/croissant_utils_test.py b/tensorflow_datasets/core/utils/croissant_utils_test.py
index 5f9bee4e579..d25ef73042f 100644
--- a/tensorflow_datasets/core/utils/croissant_utils_test.py
+++ b/tensorflow_datasets/core/utils/croissant_utils_test.py
@@ -71,7 +71,7 @@ def test_get_split_recordset():
       ),
       mlc.RecordSet(
           id='splits',
-          key='name',
+          key='splits/name',
           data_types=[mlc.DataType.SPLIT],
           fields=[
               mlc.Field(
@@ -95,7 +95,7 @@ def test_get_split_recordset_with_no_split_recordset():
   record_sets = [
       mlc.RecordSet(
           id='labels',
-          key='name',
+          key='labels/label',
           fields=[
               mlc.Field(
                   id='labels/label',
diff --git a/tensorflow_datasets/testing/test_utils.py b/tensorflow_datasets/testing/test_utils.py
index bd05e28802b..47a28a46fe6 100644
--- a/tensorflow_datasets/testing/test_utils.py
+++ b/tensorflow_datasets/testing/test_utils.py
@@ -736,6 +736,7 @@ def dummy_croissant_file(
 
   fields = [
       mlc.Field(
+          id='index',
           name='index',
           description='The sample index.',
           data_types=mlc.DataType.INTEGER,
@@ -745,6 +746,7 @@ def dummy_croissant_file(
           ),
       ),
       mlc.Field(
+          id='text',
           name='text',
           description='The dummy sample text.',
           data_types=mlc.DataType.TEXT,
@@ -758,6 +760,7 @@ def dummy_croissant_file(
   record_sets = [
       mlc.RecordSet(
           id='jsonl',
+          name='jsonl',
           description='Dummy record set.',
           fields=fields,
       )
@@ -778,6 +781,7 @@ def dummy_croissant_file(
     distribution = [
         mlc.FileObject(
             id='raw_data',
+            name='raw_data',
             description='File with the data.',
             encoding_format='application/jsonlines',
             content_url=f'data/{raw_data_filename}',