diff --git a/examples/summarization/distillation.py b/examples/summarization/distillation.py
index 529b3475..30c22326 100644
--- a/examples/summarization/distillation.py
+++ b/examples/summarization/distillation.py
@@ -11,7 +11,6 @@ from torch.nn import functional as F
 
 from lightning_base import generic_train
 from transformers import AdamW, BartConfig, BartForConditionalGeneration, T5Config, T5ForConditionalGeneration
-from transformers.modeling_bart import invert_mask
 
 
 try:
diff --git a/examples/summarization/test_summarization_examples.py b/examples/summarization/test_summarization_examples.py
index 48692a2b..3234eb46 100644
--- a/examples/summarization/test_summarization_examples.py
+++ b/examples/summarization/test_summarization_examples.py
@@ -7,7 +7,6 @@ import unittest
 from pathlib import Path
 from unittest.mock import patch
 
-import pandas as pd
 import torch
 from torch.utils.data import DataLoader
 
@@ -244,6 +243,8 @@ class TestSummarizationDistiller(unittest.TestCase):
         # self.assertEqual(len(contents), 15)
 
         metrics = pickle_load(Path(output_dir) / "metrics.pkl")
+        import pandas as pd
+
         val_df = pd.DataFrame(metrics["val"])
         train_df = pd.DataFrame(metrics["train"])
         test_df = pd.DataFrame(metrics["test"])