Multi-Cultural Commonsense Knowledge Distillation

arXiv:2402.10689v1 Announce Type: new
Abstract: Despite recent progress, large language models (LLMs) still face the challenge of appropriately reacting to the intricacies of social and cultural conventions. This paper presents 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 167K high-accuracy assertions for 30K …

abstract accuracy arxiv challenge concepts cs.cl distillation face knowledge language language models large language large language models llms methodology paper progress prompt recall social type