Literary Non-Style in LLM-Generated Text

A new study by Cory Massaro, published on arXiv, finds that LLM-generated text exhibits consistent statistical patterns in n-gram distributions that differ from human writing, revealing stylistic deficiencies. The analysis shows that higher-order n-grams correlate with semantic content, indicating that style and semantics are not cleanly separable in LLM outputs.

Literary Non-Style in LLM-Generated Text By Cory MassaroSource: arXiv cs.CL https://arxiv.org/list/cs.CL/recent arXiv:2607.17228v1 Announce Type: new Abstract: Prior work on LLM /glossary/llm -generated text has demonstrated quantitative and qualitative departures from text produced by humans. LLM-generated texts differ from human writing in style, resulting in a characteristic textual "feel," while the semantic range of LLMs is much restricted compared to that of humans. In this contribution, I note simple but consistent patterns in the statistical distribution of n-grams within LLM-generated text. Via qualitative analysis of these n-grams, I reveal deficiencies in LLM style. Because higher-order n-grams correlate to semantic content, I conclude that questions of style and semantics are not cleanly separable.Get AI news in your inbox Daily digest of what matters in AI.