2023/09/25 by Tuhin Chakrabarty, Chakrabarty, Tuhin, Philippe Laban +7 · 2 voices · 83 citations
Computer Science · Medicine · Psychology · Social Sciences · #Art #Artificial Intelligence in Healthcare and Education #Computational and Text Analysis Methods #Creative writing #Creativity #Epistemology #Flexibility (engineering) #Fluency #Literature #Management #Mathematics education #Originality #Product (mathematics) #Psychology #Quality (philosophy) #Social psychology #Test (biology) #Topic Modeling #Torrance Tests of Creative Thinking
paper · pdf · doi:10.48550/arxiv.2309.14556
published in arXiv (Cornell University) (Cornell University)
openalex publication_date 2023/09/25 · openalex created_date 2023/09/28 · openalex updated_date 2026/07/28
Researchers have argued that large language models (LLMs) exhibit high-quality writing capabilities from blogs to stories. However, evaluating objectively the creativity of a piece of writing is challenging. Inspired by the Torrance Test of Creative Thinking (TTCT), which measures creativity as a process, we use the Consensual Assessment Technique [3] and propose the Torrance Test of Creative Writing (TTCW) to evaluate creativity as a product. TTCW consists of 14 binary tests organized into the original dimensions of Fluency, Flexibility, Originality, and Elaboration. We recruit 10 creative writers and implement a human assessment of 48 stories written either by professional authors or LLMs using TTCW. Our analysis shows that LLM-generated stories pass 3-10X less TTCW tests than stories written by professionals. In addition, we explore the use of LLMs as assessors to automate the TTCW evaluation, revealing that none of the LLMs positively correlate with the expert assessments.