Exploring Chinese Humor Generation: A Study on Two-Part Allegorical Sayings
This addresses the challenge of culturally nuanced humor generation in NLP for Chinese language applications, but it is incremental as it builds on existing methods.
This paper tackled the problem of generating Chinese humor, specifically two-part allegorical sayings, using state-of-the-art language models, and found that prompting a large model was practical and effective, though it still lagged behind human creativity.
Humor, a culturally nuanced aspect of human language, poses challenges for computational understanding and generation, especially in Chinese humor, which remains relatively unexplored in the NLP community. This paper investigates the capability of state-of-the-art language models to comprehend and generate Chinese humor, specifically focusing on training them to create allegorical sayings. We employ two prominent training methods: fine-tuning a medium-sized language model and prompting a large one. Our novel fine-tuning approach incorporates fused Pinyin embeddings to consider homophones and employs contrastive learning with synthetic hard negatives to distinguish humor elements. Human-annotated results show that these models can generate humorous allegorical sayings, with prompting proving to be a practical and effective method. However, there is still room for improvement in generating allegorical sayings that match human creativity.