Update app.py

app.py

@@ -23,8 +23,8 @@ def generate_response(passage: str, question: str) -> str:
     outputs = model.generate(inputs.input_ids, max_new_tokens=150)
     
     # Decode only the generated part, skipping the prompt input
-    generated_tokens = outputs[0][inputs.input_ids.shape[-1]:]  # Ignore input tokens in the output
-    response = tokenizer.decode(generated_tokens, skip_special_tokens=True)
+    # generated_tokens = outputs[0][inputs.input_ids.shape[-1]:]  # Ignore input tokens in the output
+    response = tokenizer.batch_decode(outputs, skip_special_tokens=True)[0]
     
     return response