Update app.py

app.py

@@ -124,8 +124,9 @@ def process_and_query(files, question):
         retrieved_results = [state["sentences"][i] for i in I[0]]
 
         # Generate response based on retrieved results
-        combined_input = question + " ".join(retrieved_results)
-        inputs = generator_tokenizer(combined_input, return_tensors="pt")
+        context = " ".join(retrieved_results)
+        combined_input = f"Question: {question}\nContext: {context}"
+        inputs = generator_tokenizer(combined_input, return_tensors="pt", max_length=512, truncation=True)
         with torch.no_grad():
             generator_outputs = generator.generate(**inputs)
             generated_text = generator_tokenizer.decode(generator_outputs[0], skip_special_tokens=True)