ChatGPT had a lazy period, where it sometimes would refuse user requests if they looked like too much work. For instance, when asked to reformat a bunch of tabular data into JSON, it would do the first row, and ask the user to do the rest by themselves
PROMPT ENGINEERING
-
Weak Models Match Strong Models on Simple Prompts
By
–
In my experiment, a preference predictor was able to pick up the performance patterns of different models. One pattern is that for simple prompts, weak models can do (nearly) as well as strong models. For more challenging prompts, however, users are much more likely to prefer
-
Predicting Best AI Model Selection for User Queries
By
–
A challenge of building AI applications is choosing which model to use. What if we don’t have to? What if we can predict the best model for any prompt? Predictive human preference aims to predict which model users might prefer for a specific query. One use case is model
-
LLM Hallucinations: Chaos or Creative Genius?
By
–
LLMs "hallucinate" from nowhere… is it pure chaos… or untapped genius? Often my best ideas are similar to conventional wisdom, but tweaked a few degrees they become so much more powerful.
-
Understanding Prompts as Cognitive Seeds: Hacking AI Processes
By
–
Prompts are written as commands, but that's not what they are. They're seeds of process. One that resembles human cognition, but it's not. What happens when you start to understand how these two things are alike and how they're different? You can hack that process.
-
AI Art Courses and Tutorials
By
–
AI Art Courses:
— God of Prompt (@godofprompt) 29 février 2024
Use AI to generate art, and images for social media, blogs, and logos.
4. Midjourney Basics Course:https://t.co/VKODJiMoy3
5. Website Designing with Midjourney:https://t.co/6AFS16pCWy
6. Stable Diffusion Prompting:https://t.co/otSq0O2ah2 pic.twitter.com/UlC9Ex2kmGAI Art Courses: Use AI to generate art, and images for social media, blogs, and logos. 4. Midjourney Basics Course: https://
youtu.be/hDK8W5SLzlw 5. Website Designing with Midjourney: https://
youtu.be/5wdCev86RYE 6. Stable Diffusion Prompting: https://
youtu.be/NU2TYgfQwcA -
Generative AI Courses and Prompt Engineering
By
–
Generative AI courses:
— God of Prompt (@godofprompt) 29 février 2024
Easily learn ChatGPT and Gemini applications in real life.
And prompt engineering.
1. DataCamp: ChatGPT Business Use Cases
https://t.co/JBGKTAWmIi
2. Dignited: Google Bard Coursehttps://t.co/BxRXkrqm4r
3. Prompt Engineeringhttps://t.co/53s8FvCkxu pic.twitter.com/t4xUiyh7sCGenerative AI courses: Easily learn ChatGPT and Gemini applications in real life. And prompt engineering. 1. DataCamp: ChatGPT Business Use Cases http://
app.datacamp.com/learn/courses/
introduction-to-chatgpt
… 2. Dignited: Google Bard Course https://
youtu.be/f4LiRtjaW28 3. Prompt Engineering https://
deeplearning.ai/short-courses/
chatgpt-prompt-engineering-for-developers/
… -
Ideogram 1.0 Improves Text-to-Image Generation with Magic Prompt
By
–
3. Ideogram rolled out a new 1.0 version of its text-to-image model.
— Rowan Cheung (@rowancheung) 29 février 2024
Most notably, is the incredible details in AI-generated text and a new feature called Magic Prompt to help with creating prompts. pic.twitter.com/TCHQCxN8UZ3. Ideogram rolled out a new 1.0 version of its text-to-image model. Most notably, is the incredible details in AI-generated text and a new feature called Magic Prompt to help with creating prompts.
-
Image Generation Bias Mitigation Through Prompt Injection Techniques
By
–
But as you know, image generation is a different technology with well-understood biases. The attempts to mitigate those with prompt injection is a common approach, even if Google did it badly. https://
bloomberg.com/graphics/2023-
generative-ai-bias/
… -
LLMs Don’t Have Political Opinions, Research Shows
By
–
Asking AIs for their political opinions is a hot topic, but this paper shows it can be misleading. LLMs don’t have them: “We found that models will express diametrically opposing views depending on minimal changes in prompt phrasing or situative context” https://
arxiv.org/html/2402.1678
6v1
…