Update app.py

app.py

@@ -330,6 +330,48 @@ def format_references(
     return "\n".join(lines)
 
 
+
+def is_academic_query(message: str) -> bool:
+    """
+    粗略判断：这次学生输入是否是“学术相关 / 课程相关”问题。
+    目的：避免像 "hi"、"thanks" 这种闲聊也去查 RAG + 打 reference。
+    """
+    if not message:
+        return False
+
+    m = message.strip().lower()
+    if not m:
+        return False
+
+    # 典型闲聊词
+    smalltalk_tokens = {
+        "hi", "hello", "hey", "yo",
+        "thanks", "thank", "thank you",
+        "ok", "okay",
+        "bye", "goodbye", "see you",
+        "haha", "lol"
+    }
+
+    tokens = m.split()
+
+    # 如果全是闲聊词而且没有问号 → 视为非学术问题
+    if "?" not in m and all(t in smalltalk_tokens for t in tokens):
+        return False
+
+    # 很短、又没有问号的输入，大概率也不是严肃学术问题
+    if len(tokens) <= 2 and "?" not in m:
+        return False
+
+    # 其他情况默认当作学术/课程相关问题
+    return True
+
+
+
+
+
+
+
+
 # ================== Gradio App ==================
 with gr.Blocks(
     title="Clare – Hanbridge AI Teaching Assistant", css=CUSTOM_CSS
@@ -830,9 +872,14 @@ with gr.Blocks(
         cognitive_state = update_cognitive_state_from_message(message, cognitive_state)
 
         # RAG 始终基于 (预加载 Module10 + 可选上传)
-        rag_context_text, rag_used_chunks = retrieve_relevant_chunks(
-            message, rag_chunks or []
-        )
+        # 只对“学术/课程相关”问题调用 RAG；闲聊不查文档，也不打 reference
+        if is_academic_query(message):
+            rag_context_text, rag_used_chunks = retrieve_relevant_chunks(
+                message, rag_chunks or []
+            )
+        else:
+            rag_context_text, rag_used_chunks = "", []
+
 
         # 计时
         start_ts = time.time()
@@ -851,15 +898,18 @@ with gr.Blocks(
         end_ts = time.time()
         latency_ms = (end_ts - start_ts) * 1000.0
 
-        # === 在这里附上 References ===
-        ref_text = format_references(rag_used_chunks)
+        # === 在这里附上 References（只在学术问题上附） ===
+        ref_text = ""
+        if is_academic_query(message) and rag_used_chunks:
+            ref_text = format_references(rag_used_chunks)
+
         if ref_text and new_history:
             last_user, last_assistant = new_history[-1]
             if "References (RAG context used):" not in (last_assistant or ""):
                 last_assistant = f"{last_assistant}\n\n{ref_text}"
                 new_history[-1] = [last_user, last_assistant]
                 answer = last_assistant  # 同步给日志
-        # ==============================
+        # ============================================
 
         # 日志
         student_id = user_id_val or "ANON"