In this Issue:

1. An overview of active prompting.

2. A review of the original active prompting paper by researchers from the University of Toronto.

3. A walkthrough Microsoft Guidance framework for language model programming.

💡 ML Concept of the Day: What is Active Prompting?

Active prompting is one of the most novel techniques used to determine the effectiness of certain prompts in order to accomplish specific tasks.

It is widely acknowledged that the effective formulation of task-specific prompts plays a crucial role in enabling LLMs to generate high-quality answers. Particularly, an efficacious method for addressing complex question-and-answer tasks involves example-based prompting using chain-of-thought (CoT) reasoning, which significantly enhances the performance of LLMs. However, current CoT techniques rely on a predetermined collection of human-annotated instances, which may not always be the most suitable examples for diverse tasks...