diff --git a/pred.py b/pred.py
index 1c20cba9..5f0f5d48 100644
--- a/pred.py
+++ b/pred.py
@@ -53,7 +53,7 @@ def query_llm(prompt, model, tokenizer, client=None, temperature=0.5, max_new_to
             time.sleep(1)
     else:
         print("Max tries. Failed.")
-        return ''
+        return None
 
 def extract_answer(response):
     response = response.replace('*', '')
@@ -95,14 +95,14 @@ def get_pred(data, args, fout):
             output = query_llm(prompt, model, tokenizer, client, temperature=0.1, max_new_tokens=1024)
         else:
             output = query_llm(prompt, model, tokenizer, client, temperature=0.1, max_new_tokens=128)
-        if output == '':
+        if output == None:
             continue
         if args.cot: # extract answer
             response = output.strip()
             item['response_cot'] = response
             prompt = template_0shot_cot_ans.replace('$DOC$', context.strip()).replace('$Q$', item['question'].strip()).replace('$C_A$', item['choice_A'].strip()).replace('$C_B$', item['choice_B'].strip()).replace('$C_C$', item['choice_C'].strip()).replace('$C_D$', item['choice_D'].strip()).replace('$COT$', response)
             output = query_llm(prompt, model, tokenizer, client, temperature=0.1, max_new_tokens=128)
-            if output == '':
+            if output == None:
                 continue
         response = output.strip()
         item['response'] = response