Researchers explored the capacity of large language models (LLMs) to generate infinite poems, finding that while AI can produce interesting word pairings, it falls short of human-level creativity. Two experiments were conducted: one involving manual prompting and co-creation, and another utilizing an LLM with semantic similarity to achieve automated output. The findings suggest that current LLMs have limitations in semantic and quantitative poetic generation, but incorporating AI workflows could lead to the revival and advancement of computational poetry. AI
IMPACT Highlights limitations in current LLMs for creative tasks, suggesting AI integration may advance computational poetry.
RANK_REASON The cluster contains an academic paper detailing experiments on LLM capabilities in poetry generation. [lever_c_demoted from research: ic=1 ai=1.0]
- alphaXiv
- arXiv
- CatalyzeX Code Finder for Papers
- CORE Recommender
- DagsHub
- Gotit.pub
- Hugging Face
- Influence Flower
- ScienceCast
- Vincent Cellucci
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →