[Coati] Refine prompt for better inference (#6117)

* refine prompt

* update prompt

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

Author: TongLi3701

applications/ColossalChat/coati/reasoner/guided_search/mcts.py

@@ -58,7 +58,7 @@ def initialization(self):
         """
         Root Initiation.
         """
-        # Dummy answer as root.
+        # Simple answer as root. You can also use negative response such as "I do not know" as a response.
         base_answer = self.sample_base_answer()
         self.root = MCTSNode(answer=base_answer)
         self.self_evaluate(self.root)
@@ -190,7 +190,7 @@ def sample_base_answer(self):
             messages=[
                 {
                     "role": "system",
-                    "content": "The user will provide a problem. Solve the problem. The response should begin with [reasoning process]...[Verification]... and end with [Final Answer]. \nThe answer is [answer] \n#### [answer].",
+                    "content": self.cfg.base_system_prompt,
                 },
                 {
                     "role": "user",

applications/ColossalChat/coati/reasoner/guided_search/prompt_store/base.py

@@ -5,6 +5,7 @@ class PromptCFG(BaseModel):
     model: str
     base_url: str
     max_tokens: int = 4096
+    base_system_prompt: str
     critic_system_prompt: str
     refine_system_prompt: str
     evaluate_system_prompt: str

applications/ColossalChat/coati/reasoner/guided_search/prompt_store/qwen.py

@@ -7,14 +7,16 @@
 Qwen32B_prompt_CFG = PromptCFG(
     base_url="http://0.0.0.0:8008/v1",
     model="Qwen2.5-32B-Instruct",
-    critic_system_prompt="Provide a detailed and constructive critique to improve the answer. "
-    "Highlight specific areas that need refinement or correction.",
+    base_system_prompt="The user will present a problem. Analyze and solve the problem in the following structure:\n"
+    "Begin with [Reasoning Process] to explain the approach. \n Proceed with [Verification] to confirm the solution. \n Conclude with [Final Answer] in the format: 'Answer: [answer]'",
+    critic_system_prompt="Provide a detailed and constructive critique of the answer, focusing on ways to improve its clarity, accuracy, and relevance."
+    "Highlight specific areas that need refinement or correction, and offer concrete suggestions for enhancing the overall quality and effectiveness of the response.",
     refine_system_prompt="""# Instruction
                             Refine the answer based on the critique. The response should begin with [reasoning process]...[Verification]... and end with [Final Answer].
                          """,
     evaluate_system_prompt=(
-        "Analyze this answer strictly and critic, provide a reward score between -100 and 100 for the answer quality, using very strict standards. "
-        "Do not give a full score above 95. Make sure the reward score is an integer. "
-        "Return *ONLY* the score."
+        "Critically analyze this answer and provide a reward score between -100 and 100 based on strict standards."
+        "The score should clearly reflect the quality of the answer."
+        "Make sure the reward score is an integer. You should only return the score. If the score is greater than 95, return 95."
     ),
 )