diff --git a/opencompass/datasets/generic.py b/opencompass/datasets/generic.py
index 5570d317..07b6a0bb 100644
--- a/opencompass/datasets/generic.py
+++ b/opencompass/datasets/generic.py
@@ -37,6 +37,7 @@ def get_final_results(judged_answers,
     is_correct = is_correct_count / count
     is_incorrect = is_incorrect_count / count
     is_given_attempted = is_correct + is_incorrect
+    loose_accuracy = is_correct / count
     accuracy_given_attempted = (is_correct / is_given_attempted
                                 if is_given_attempted > 0 else 0)
     attempted_judge_ratio = attempted_judge_count / count
@@ -45,7 +46,8 @@ def get_final_results(judged_answers,
           (accuracy_given_attempted + is_correct) if
           (accuracy_given_attempted + is_correct) > 0 else 0)
     result = {
-        metric_name: accuracy_given_attempted * 100,
+        metric_name: loose_accuracy * 100,
+        f'{metric_name}_given_attempted': accuracy_given_attempted * 100,
         'f1': f1,
         'attempted_ratio': attempted_judge_ratio * 100,
         'correct_count': is_correct_count,