arXiv cs.CL· Kamer Cekini, Elisabetta Biondi, Chiara Boldrini, Andrea Passarella, Marco Conti·· 12 小时前AI 评分31
研究：LLM 改写文本的阅读时间与认知成本对比
The Cognitive Cost of LLM Writing: A Reading-Time Comparison with Human Text
AI 导读
研究者提出 RoBERTa-RT 模型，基于人类眼动数据预测词级阅读时间，并用 496 篇人类文章及其 LLM 改写文本做配对比较。结果显示 LLM 改写文本更短、预测总阅读时间略低，但每词预测阅读时间更高，伴随更长且更低频的词、更高词汇多样性和更高上下文可预测性。研究认为 LLM 写作可能重新分配阅读负担：减少需处理的文本量，却提高单个词的预测加工成本。
正文 · 原文
Abstract:Although linguistic differences between human-written and Large Language Model (LLM)-generated texts have been extensively studied, their implications for the cognitive effort involved in reading remain less explored. In psycholinguistics, reading time is commonly used as an indirect measure of cognitive processing effort. Building on this perspective, we investigate whether human-written texts and their LLM-generated rewrites exhibit differences in predicted reading time that may reflect differences in processing demands for readers. To this end, we introduce RoBERTa-RT, a model trained on human eye-tracking data to predict word-level reading times. We evaluate the model on both internal and external test sets, construct a paired dataset of 496 human-written articles and their LLM-generated rewrites, and apply RoBERTa-RT to compare their predicted reading times. Our findings show that LLM-generated rewrites tend to be shorter and have slightly lower predicted total reading times, while exhibiting higher predicted reading times per word. These differences are associated with systematic linguistic changes, including longer and less frequent words, greater lexical diversity, and higher contextual predictability. From a cognitive perspective, our findings suggest that LLM-generated writing may redistribute reading demands: reducing the amount of text to process while increasing the predicted processing effort associated with individual words. This highlights the importance of considering not only the linguistic characteristics of machine-generated content, but also its potential implications for human cognitive processing.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.23715 [cs.CL] |
| (or arXiv:2610.23715v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.23715 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Kamer Cekini [view email]
[v1]
Fri, 16 Oct 2026 16:35:41 UTC (95 KB)
来源：arXiv cs.CL · arxiv.org