New metric for joke humor shows promise despite poor prediction accuracy

Researchers have explored the semantic distance of double meanings in jokes, building on prior work that suggested humor increases with the frequency and rarity of associated terms, as well as the ambiguity and meaning distance between concepts. Using two embedding models, OpenAI text-embedding-3-small and MiniLM all-MiniLM-L6-v2, on joke datasets, the study found that models trained on proposed metrics performed poorly in predicting humor ratings. However, a newly introduced metric, symmetry, showed a consistent association with higher-rated jokes, indicating it might be a necessary component of humor. AI

IMPACT Introduces a new metric for humor analysis, potentially improving AI's understanding of wordplay and jokes.

RANK_REASON Academic paper presenting new metrics for computational humor research. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

• Expunations
• Hugging Face
• JokeJudger
• Matthews
• MiniLM all-MiniLM-L6-v2
• OpenAI
• Petrović
• rJokes
• text-embedding-3-small
• Winters
• Word2Vector

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →