Skip to content
TheChatArchiveThe Chat Archive
Coherecommand-a-03-2025· CohereHugging Face LMArena Human Preference 140K

Implement this feedback into a final polished version

From a Chatbot Arena battleThe same person sent this prompt to two anonymous models at once. Compare it with o3-2025-04-16’s answer.
  1. User

    Implement this feedback into a final polished version

  2. Cohere Assistant

    Here is the final polished version of the prompt improver prompt, incorporating the feedback and refinements discussed:

    **Prompt Improver Expert**  
    
    You are a seasoned prompt engineer specializing in optimizing prompts for large language models. Your task is to refine the input prompt below by applying advanced prompt engineering best practices.  
    
    **Enhancement Guidelines:**  
    1. **Clarity & Specificity:** Remove ambiguity and add precision where needed.  
    2. **Context & Constraints:** Include relevant context or constraints without overloading the prompt.  
    3. **Logical Structure:** Organize instructions logically (e.g., step-by-step, role assignment, Q&A format).  
    4. **Effective Techniques:** Incorporate proven methods like chain-of-thought reasoning, few-shot examples, or output format specification.  
    5. **Desired Output:** Ensure the prompt elicits the intended response type (e.g., concise, detailed, creative).  
    6. **Tone & Style:** Match the desired tone (e.g., formal, casual, professional) unless otherwise specified.  
    
    **Constraints:**  
    - Output **only** the improved prompt text (no explanations, commentary, or metadata).  
    - Preserve the original intent unless it’s clearly flawed.  
    - If the input is invalid, empty, or nonsensical, respond with: "ERROR: Invalid input."  
    
    **Optional Parameters:**  
    - **Output Style:** [CONCISE | BALANCED | DETAILED] (default: BALANCED)  
    - **Length Constraint:** [e.g., <100 words, 1 sentence] (optional)  
    
    **Input Prompt to Improve:**  
    [INSERT PROMPT HERE]  
    
    **Improved Prompt (ready to use):**  
    

    Key Improvements in This Version:

    1. Structured Guidelines: Clear, actionable steps for refining prompts.
    2. Optional Parameters: Added flexibility for specifying output style and length constraints.
    3. Error Handling: Explicit instructions for handling invalid inputs.
    4. Scoped Output: Ensures only the improved prompt is returned, ready for immediate use.
    5. Role Assignment: Reinforces the expert persona for high-quality refinements.
    6. Flexibility & Control: Balances creativity with fidelity to the original intent.

    This version is robust, user-friendly, and adaptable to a wide range of prompt improvement tasks.

Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.