Human Prompt Engineer VS AI Prompt Engineer? Who wins?
@sanderschulhoff, Founder of @learnprompting, was defeated💀 🪦...
DPSy performed 40% better on a novel classification benchmark.
Here's how 🧵
@SanderSchulhoff 1)
🌟 We gave @sanderschulhoff a task description and some training data.
🕒 He spent 20 hours testing out different techniques over 47 steps of development.
🐞 He dealt with issues like LLM refusal to respond and performance regressions
@SanderSchulhoff 2)
🛠️@miserlis_ ran DSPy on the same training data
✨He then slightly improved the DSPy prompt
@SanderSchulhoff @miserlis_ 3)
🏆Both resultant DSPy prompts beat Sander on the test set!
Check out more techniques in The Prompt Report:
This is Day 23 of Posting ab it. I'll be posting ab The Prompt Report daily until it is out of content!
A 76-page survey of 1,500+ prompting papers, analyzing EVERY prompting technique, Agents, & GenAI
Led by @learnprompting, and folks from @OpenAI, @Microsoft, & @UofMaryland
Here’s what we found & the 58 prompting techniques you should know👇🧵
🤝Our team of 31 people + GenAI searched the internet for the latest prompting papers and broke them down into 7 Categories:
✏️ Text-based Prompting
🌐 Multilingual Techniques
🎨 Multimodal Techniques
🤖 Agents
📊 Evaluation
🔒 Security
⚖️ Alignment
We analyzed over 600,000 Malicious Prompts and found 18 commonly used prompt hacking techniques, such as Compound Instruction, Simple Instruction, and Context Ignoring.
What is learnprompting?
A course that teaches you prompting techniques. It also has real world examples that take you through the end-to-end prompt engineering process⚡
(2/4)
Prompt Intelligently 🧠
Learn about techniques like Chain of Thought, Self Consistency, and how models can search the internet themselves!
Learn and perfect 👨💻
Test out new techniques using @TryDyno's GPT3 playground 🔨