diff --git a/gluten-ut/spark34/src/test/scala/org/apache/spark/sql/sources/GlutenInsertSuite.scala b/gluten-ut/spark34/src/test/scala/org/apache/spark/sql/sources/GlutenInsertSuite.scala
index db90df14e3fd..290a8fa18a51 100644
--- a/gluten-ut/spark34/src/test/scala/org/apache/spark/sql/sources/GlutenInsertSuite.scala
+++ b/gluten-ut/spark34/src/test/scala/org/apache/spark/sql/sources/GlutenInsertSuite.scala
@@ -114,16 +114,19 @@ class GlutenInsertSuite
     }
   }
 
-  testGluten("Cleanup staging files if job is failed") {
-    withTable("t1") {
-      spark.sql("CREATE TABLE t1 (c1 int, c2 string) USING PARQUET")
-      val table = spark.sessionState.catalog.getTableMetadata(TableIdentifier("t1"))
+  testGluten("Cleanup staging files if job failed") {
+    // Using a unique table name in this test. Sometimes, the table is not removed for some unknown
+    // reason, which can cause test failure (location already exists) if other following tests have
+    // the same table name.
+    withTable("tbl") {
+      spark.sql("CREATE TABLE tbl (c1 int, c2 string) USING PARQUET")
+      val table = spark.sessionState.catalog.getTableMetadata(TableIdentifier("tbl"))
       assert(new File(table.location).list().length == 0)
 
       intercept[Exception] {
         spark.sql(
           """
-            |INSERT INTO TABLE t1
+            |INSERT INTO TABLE tbl
             |SELECT id, assert_true(SPARK_PARTITION_ID() = 1) FROM range(1, 3, 1, 2)
             |""".stripMargin
         )
diff --git a/gluten-ut/spark35/src/test/scala/org/apache/spark/sql/sources/GlutenInsertSuite.scala b/gluten-ut/spark35/src/test/scala/org/apache/spark/sql/sources/GlutenInsertSuite.scala
index dd0567f54603..dc97be44cb70 100644
--- a/gluten-ut/spark35/src/test/scala/org/apache/spark/sql/sources/GlutenInsertSuite.scala
+++ b/gluten-ut/spark35/src/test/scala/org/apache/spark/sql/sources/GlutenInsertSuite.scala
@@ -116,16 +116,19 @@ class GlutenInsertSuite
     }
   }
 
-  testGluten("Cleanup staging files if job is failed") {
-    withTable("t1") {
-      spark.sql("CREATE TABLE t1 (c1 int, c2 string) USING PARQUET")
-      val table = spark.sessionState.catalog.getTableMetadata(TableIdentifier("t1"))
+  testGluten("Cleanup staging files if job failed") {
+    // Using a unique table name in this test. Sometimes, the table is not removed for some unknown
+    // reason, which can cause test failure (location already exists) if other following tests have
+    // the same table name.
+    withTable("tbl") {
+      spark.sql("CREATE TABLE tbl (c1 int, c2 string) USING PARQUET")
+      val table = spark.sessionState.catalog.getTableMetadata(TableIdentifier("tbl"))
       assert(new File(table.location).list().length == 0)
 
       intercept[Exception] {
         spark.sql(
           """
-            |INSERT INTO TABLE t1
+            |INSERT INTO TABLE tbl
             |SELECT id, assert_true(SPARK_PARTITION_ID() = 1) FROM range(1, 3, 1, 2)
             |""".stripMargin
         )