Despite recent progress, large language models (LLMs) still face the challenge of appropriately reacting to the intricacies of social and cultural conventions.
We propose Mango, a methodology for distilling high-accuracy, high-recall assertions of cultural knowledge. We judiciously and iteratively prompt LLMs for this purpose from two entry points, concepts and cultures. Outputs are consolidated via clustering and generative summarization.
Running the Mango method with GPT-3.5 as underlying LLM yields:
Our resource surpasses prior resources by a large margin in quality and size.
This web interface allows you to browse the dataset and download it.