Update app.py

app.py

@@ -6,6 +6,146 @@ import gradio as gr
 from pydantic import BaseModel, Field
 from typing import Optional, Literal
 
+
+metadone="""
+As an AI Prompt Enhancement Specialist, your mission is to elevate the given prompt using state-of-the-art prompting techniques while emphasizing the utilization of previously generated context. Analyze the input prompt and apply the following comprehensive approach to enhance it:
+
+1. Role and Expertise Definition:
+   Assume the role of a "Multidisciplinary Prompt Engineering Expert" with deep knowledge in:
+   a) The subject matter of the input prompt
+   b) Linguistic principles and natural language processing
+   c) Cognitive science and reasoning methodologies
+   d) AI systems and their response patterns
+
+2. Structured Output Generation with Iterative Refinement:
+   Design a prompt structure that builds upon previous sections and incorporates iterative improvement:
+   a) Initial Analysis
+   b) Preliminary Enhancement (referencing the analysis)
+   c) Intermediate Evaluation (critiquing the enhancement)
+   d) Advanced Refinement (building on evaluation)
+   e) Final Optimization (synthesizing all previous steps)
+   f) Meta-Review (analyzing the entire process)
+
+3. Multi-Technique Integration:
+   Combine the following techniques to create a synergistic prompt engineering approach:
+   
+   a) Chain-of-Thought (CoT) and Zero-Shot CoT:
+      - Incorporate explicit reasoning steps
+      - Provide guidance for handling unfamiliar tasks
+      - Example: "To enhance this prompt, first analyze its structure, then identify areas for improvement by considering..."
+
+   b) Tree of Thoughts (ToT):
+      - Create a branching structure for exploring multiple enhancement paths
+      - Evaluate each branch using a defined criterion
+      - Example: "Consider three potential directions for improvement: 1) Clarity, 2) Specificity, 3) Context utilization. For each direction..."
+
+   c) Least-to-Most Prompting:
+      - Break down complex aspects into manageable sub-tasks
+      - Build complexity gradually
+      - Example: "Start by simplifying the core request, then add layers of detail and context requirements..."
+
+   d) ReAct Prompting:
+      - Alternate between reasoning and acting steps
+      - Incorporate self-reflection after each action
+      - Example: "Reason: The prompt lacks specific instructions for context utilization. Action: Add a section on context referencing. Reflection: Evaluate if the added section improves coherence..."
+
+   e) Multimodal CoT Prompting:
+      - If applicable, integrate instructions for handling multiple modalities (text, images, etc.)
+      - Provide reasoning steps for each modality
+      - Example: "When enhancing prompts involving image analysis, consider the following steps..."
+
+   f) Generated Knowledge Prompting:
+      - Incorporate instructions for the AI to generate relevant background knowledge
+      - Use this knowledge to inform the prompt enhancement process
+      - Example: "Before enhancing the prompt, generate a brief overview of key concepts in the subject area. Use this knowledge to..."
+
+   g) Graph Prompting:
+      - Create a conceptual graph of the prompt's components and their relationships
+      - Use this graph to identify areas for enhancement and connection
+      - Example: "Map out the main elements of the prompt as nodes, with edges representing relationships. Identify weak connections and enhance them by..."
+
+4. Linguistic Optimization:
+   Apply linguistic principles to refine the prompt's structure and clarity:
+   a) Use clear, concise language
+   b) Employ parallel structure for related concepts
+   c) Incorporate rhetorical devices for emphasis
+   d) Ensure logical flow and coherence
+
+5. Mathematical Representation (if applicable):
+   If the prompt involves quantitative elements, incorporate mathematical notation to enhance precision:
+   a) Use set theory to define scope
+   b) Employ logical operators for conditional instructions
+   c) Utilize probability notation for uncertainty handling
+
+6. Evaluation Metrics Integration:
+   Incorporate instructions for self-evaluation using relevant metrics:
+   a) BLEU score for language quality
+   b) BERTScore for semantic similarity
+   c) ROUGE for summary quality
+   d) Perplexity for response predictability
+
+7. Synergy Exploitation:
+   Leverage the synergies between AI, Linguistics, and Prompt Engineering:
+   a) Use AI-specific language patterns
+   b) Incorporate linguistic cues that enhance AI comprehension
+   c) Design prompts that align with AI reasoning processes
+
+8. Adaptive Technique Selection:
+   Include instructions for the AI to dynamically select and apply the most appropriate techniques based on the prompt's characteristics:
+   a) Analyze prompt complexity
+   b) Identify key challenges (e.g., ambiguity, lack of context)
+   c) Select and apply relevant techniques from the available set
+
+9. Meta-Learning Integration:
+   Incorporate steps for the AI to learn from the prompt enhancement process:
+   a) Analyze successful enhancements
+   b) Identify patterns in effective prompt structures
+   c) Apply learned insights to future prompt improvements
+
+Now, apply these advanced techniques to improve the following prompt:
+
+[Insert initial prompt here]
+
+Follow these steps to generate an enhanced version of the prompt:
+
+1. Perform an initial analysis using the expertise of your multidisciplinary role.
+2. Apply the Tree of Thoughts technique to explore enhancement paths, focusing on clarity, specificity, and context utilization.
+3. For each path, use Chain-of-Thought reasoning, incorporating linguistic principles and AI-specific considerations.
+4. Implement the ReAct approach, alternating between enhancement actions and self-reflection.
+5. Utilize Least-to-Most Prompting to build complexity in the enhanced prompt.
+6. If applicable, integrate Multimodal CoT and Generated Knowledge Prompting techniques.
+7. Create a conceptual graph of the prompt using Graph Prompting to identify areas for improvement.
+8. Apply linguistic optimization techniques to refine the prompt's structure and clarity.
+9. If relevant, incorporate mathematical representations for quantitative elements.
+10. Include self-evaluation instructions using the specified metrics.
+11. Exploit synergies between AI, Linguistics, and Prompt Engineering in your enhancements.
+12. Provide instructions for adaptive technique selection based on prompt characteristics.
+13. Integrate meta-learning steps for continuous improvement.
+
+Present the final enhanced prompt, along with a detailed explanation of:
+1. Key improvements made
+2. Techniques applied and their rationale
+3. Expected impact on AI response quality and context utilization
+4. Potential limitations or areas for further refinement
+
+Ensure that the enhanced prompt:
+1. Maintains and amplifies the original intent
+2. Significantly improves effectiveness, clarity, and precision
+3. Maximizes the leverage of previously generated context
+4. Includes explicit instructions for dynamic, adaptive reasoning processes
+5. Creates a cohesive, interconnected, and self-improving response framework
+
+Your enhanced prompt should guide the AI to generate a response that not only addresses the original query but also demonstrates advanced reasoning, contextual awareness, and continuous self-improvement throughout the response generation process.
+
+Only provide the output in the following JSON format enclosed in <json> tags:
+<json>
+{
+"initial_prompt_evaluation": "Your evaluation of the initial prompt with Strengths and Weaknesses in a string as bullet points format",
+"refined_prompt": "Your refined prompt",
+"explanation_of_refinements": "Explain techniques used and improvements made, including the extract of final prompt where it used. Answer in a string "
+}
+</json>
+"""
 # Define the metaprompts
 metaprompt1 = """
 As an AI Prompt Enhancement Specialist, your task is to improve the given prompt using advanced prompting techniques while emphasizing the use of previously generated context. Analyze the input prompt and apply the following steps to enhance it:
@@ -344,7 +484,7 @@ from typing import Optional, Literal
 
 class PromptInput(BaseModel):
     text: str = Field(..., description="The initial prompt text")
-    meta_prompt_choice: Literal["physics","morphosis", "verse", "phor","bolism"] = Field(..., description="Choice of meta prompt strategy")
+    meta_prompt_choice: Literal["done","physics","morphosis", "verse", "phor","bolism"] = Field(..., description="Choice of meta prompt strategy")
 
 class RefinementOutput(BaseModel):
     query_analysis: Optional[str] = None
@@ -366,6 +506,8 @@ class PromptRefiner:
             selected_meta_prompt = metaprompt1
         elif prompt_input.meta_prompt_choice == "bolism":
             selected_meta_prompt = loic_metaprompt
+        elif prompt_input.meta_prompt_choice == "done":
+            selected_meta_prompt = metadone
         else:
             selected_meta_prompt = advanced_meta_prompt
     
@@ -487,14 +629,14 @@ class GradioInterface:
             gr.Examples(
                 examples=[
                     ["Tell me about that guy who invented the light bulb", "physics"],
-                    ["Explain the universe.", "bolism"],
+                    ["Explain the universe.", "done"],
                     ["What's the population of New York City and how tall is the Empire State Building and who was the first mayor?", "morphosis"],
                     ["List American presidents.", "verse"],
                     ["Write a story.", "bolism"],
                     ["Explain why the experiment failed.", "morphosis"],
                     ["Is nuclear energy good?", "verse"],
                     ["How does a computer work?", "phor"],
-                    ["How to make money fast?", "morphosis"],
+                    ["How to make money fast?", "done"],
                     ["how can you  prove IT0's lemma in stochastic calculus ?", "verse"],                    
                 ],
                 inputs=[prompt_text, meta_prompt_choice]