Standardizing the Measurement of Text Diversity: A Tool and a Comparative Analysis of Scores

arXiv:2403.00553v1 Announce Type: new
Abstract: The diversity across outputs generated by large language models shapes the perception of their quality and utility. Prompt leaks, templated answer structure, and canned responses across different interactions are readily noticed by people, but there is no standard score to measure this aspect of model behavior. In this work we empirically investigate diversity scores on English texts. We find that computationally efficient compression algorithms capture information similar to what is measured by slow to compute …

abstract analysis arxiv comparative analysis cs.cl diversity generated interactions language language models large language large language models leaks measurement people perception prompt quality responses standard text tool type utility