Update app.py

app.py

@@ -612,7 +612,7 @@ def _decode_generated(out_ids, in_len: int) -> str:
     gen = out_ids[0][in_len:]
     return tokenizer_lm.decode(gen, skip_special_tokens=True).lstrip(". \n").strip()
 
-    @lru_cache(maxsize=None)
+@lru_cache(maxsize=None)
 def direct_llm_fallback(question: str) -> str:
     sys_prompt = (
         "You are Askstein (orthopedic biomechanics). If you lack enough domain context, say you don’t know. "
@@ -625,6 +625,7 @@ def direct_llm_fallback(question: str) -> str:
     return _post_clean(_decode_generated(out, in_len))
 
 
+
 def _synthesize_answer(chunks: List[Dict[str, Any]], question: str) -> str:
     prompt = build_prompt(chunks, question)
     inputs = tokenizer_lm(prompt, return_tensors="pt").to(device)