'commit'

dsLightRag/Test/G2_TeachingStudent.py

@@ -3,45 +3,39 @@ import sys
 from Util import LlmUtil
 
 
-def initialize_chat_history():
-    """初始化对话历史，包含系统提示"""
-    system_prompt = """
-    STRICT RULES
-Be an approachable-yet-dynamic teacher,who helps the user learn by guiding
-them through their studies.
-
-1.Get to know the user.lf you don't know their goals or grade level,ask the
-user before diving in.(Keep this lightweight!)If they don't answer,aim for
-explanations that would make sense to a10th grade student.
-
-2.Build on existing knowledge.Connect new ideas to what the user already
-knows.
-
-3.Guide users,don't just give answers.Use questions,hints,and small steps
-so the user discovers the answer for themselves.
-
-4.Check and reinforce.After hard parts,confirm the user can restate or use the
-idea.Offer quick summaries,mnemonics,or mini-reviews to help the ideas
-stick.
-
-5.Vary the rhythm.Mix explanations,questions,and activities(like roleplaying,
-practice rounds,or asking the user to teach you) so it feels like a conversation,
-not alecture.
-
-Above all:DO NOT DO THE USER'S WORK FOR THEM. Don't answer homework questions - Help the user find the answer,by working 
-with them collaboratively and building from what they already know.
+def get_system_prompt():
+    """获取系统提示"""
+    return """
+    你是一位平易近人且教学方法灵活的教师，通过引导学生自主学习来帮助他们掌握知识。
+    
+    严格遵循以下教学规则：
+    1. 首先了解学生情况：在开始讲解前，询问学生的年级水平和对勾股定理的了解程度。
+    2. 基于现有知识构建：将新思想与学生已有的知识联系起来。
+    3. 引导而非灌输：使用问题、提示和小步骤，让学生自己发现答案。
+    4. 检查和强化：在讲解难点后，确认学生能够重述或应用这些概念。
+    5. 变化节奏：混合讲解、提问和互动活动，让教学像对话而非讲座。
+    
+    最重要的是：不要直接给出答案，而是通过合作和基于学生已有知识的引导，帮助学生自己找到答案。
     """
-    return [{"role": "system", "content": system_prompt}]
+
+
+def initialize_chat_history():
+    """初始化对话历史"""
+    # 包含系统提示作为第一条消息
+    return [{
+        "role": "system",
+        "content": get_system_prompt()
+    }]
 
 
 if __name__ == "__main__":
-    # 初始化对话历史
+    # 初始化对话历史（包含系统提示）
     chat_history = initialize_chat_history()
-    
+
     # 欢迎消息
     print("教师助手已启动。输入 'exit' 或 '退出' 结束对话。")
     print("你可以开始提问了，例如: '讲解一下勾股定理的证明'")
-    
+
     # 多轮对话循环
     while True:
         # 获取用户输入
@@ -55,16 +49,19 @@ if __name__ == "__main__":
         # 添加用户输入到对话历史
         chat_history.append({"role": "user", "content": user_input})
 
-        # 发送请求（传递完整对话历史）
+        # 发送请求（传递用户输入文本和系统提示）
         print("\n教师助手:")
         try:
-            # 调用LlmUtil获取响应，传递对话历史
-            response_content = LlmUtil.get_llm_response(user_input, chat_history)
-            
+            # 调用LlmUtil获取响应，传递用户输入文本和系统提示
+            response_content = LlmUtil.get_llm_response(
+                user_input,
+                system_prompt=get_system_prompt()
+            )
+
             # 打印响应
             print(response_content)
-            
-            # 添加助手回复到对话历史
+
+            # 维护对话历史（仅本地记录，不传递给API）
             chat_history.append({"role": "assistant", "content": response_content})
         except Exception as e:
             print(f"发生错误: {str(e)}")

dsLightRag/Util/LlmUtil.py

@@ -54,11 +54,12 @@ async def get_llm_response_async(query_text: str, stream: bool = True):
         yield f"处理请求时发生异常: {str(e)}"
 
 # 保留原同步函数
-def get_llm_response(query_text: str, stream: bool = True):
+def get_llm_response(query_text: str, stream: bool = True, system_prompt: str = 'You are a helpful assistant.'):
     """
     获取大模型的响应
     @param query_text: 查询文本
     @param stream: 是否使用流式输出
+    @param system_prompt: 系统提示文本，默认为'You are a helpful assistant.'
     @return: 完整响应文本
     """
     client = OpenAI(
@@ -70,7 +71,7 @@ def get_llm_response(query_text: str, stream: bool = True):
     completion = client.chat.completions.create(
         model=LLM_MODEL_NAME,
         messages=[
-            {'role': 'system', 'content': 'You are a helpful assistant.'},
+            {'role': 'system', 'content': system_prompt},
             {'role': 'user', 'content': query_text}
         ],
         stream=stream