通过作者语言模型的困惑来归因作者的身份
Weihang Huang1, Akira Murakami1, Jack Grieve1
1Department of Linguistics and Communication, University of Birmingham, Birmingham, West Midlands, United Kingdom.
PloS one
|July 3, 2025
概括
本研究引入了一种使用微调的大型语言模型 (LLM) 的新作者归因方法. 该方法通过使用作者语言模型 (ALM) 来测量文档可预测性,从而优于现有的方法.
科学领域:
- 计算语言学 计算语言学
- 数字法医学数字法医学
- 造型测量仪使用的方法
背景情况:
- 作者归属传统上依赖于定量造型方法.
- 大型语言模型 (LLM) 的出现为作者分析提供了新的机会.
- 现有的方法可能无法充分利用现代AI在文本分析方面的能力.
研究的目的:
- 开发和评估一种新的作者归属技术,利用精心调整的法学士.
- 将拟议的基于LLM的方法的性能与当前最先进的方法进行比较.
- 为了调查有助于成功的作者归因的语言特征.
主要方法:
- 在个人作者写作样本上对LLM进行进一步的预训,以创建作者语言模型 (ALM).
- 将被质疑的文档分配给ALM产生最低困惑度 (最高可预测性) 的作者.
- 分析词级可预测性,检查语言模式驱动归因决策.
主要成果:
- 拟议的基于LLM的作者归因方法在标准数据集上达到或超过了最新的性能.
- 该方法允许直接检查影响归因的语言模式.
- 发现内容词比功能词带有更多的作者信息,挑战了先前的假设.
结论:
- 精心调整的LLM提供了一种强大而有效的作者的归因方法.
- 该方法为归因驱动因素提供了可解释的见解.
- 这些发现建议重新评估不同词类在造型学中的作用.
相关概念视频
Pleiotropy
41.2K
Pleiotropy is the phenomenon in which a single gene impacts multiple, seemingly unrelated phenotypic traits. For example, defects in the SOX10 gene cause Waardenburg Syndrome Type 4, or WS4, which can cause defects in pigmentation, hearing impairments, and an absence of intestinal contractions necessary for elimination. This diversity of phenotypes results from the expression pattern of SOX10 in early embryonic and fetal development. SOX10 is found in neural crest cells that form melanocytes,...
41.2K
Genetic Lingo
104.9K
Overview
104.9K
Probability Laws
41.9K
Overview
41.9K
Determination
19.2K
During embryogenesis, cells become progressively committed to different fates through a two-step process: specification followed by determination. Specification is demonstrated by removing a segment of an early embryo, “neutrally” culturing the tissue in vitro—for example, in a petri dish with simple medium—and then observing the derivatives. If the cultured region gives rise to cell types that it would normally generate in the embryo, this means that it is specified. In...
19.2K
Proofreading
55.2K
Overview
55.2K
Translation
144.0K
Lesson: Translation
Translation is the process of synthesizing proteins from the genetic information carried by messenger RNA (mRNA). Following transcription, it constitutes the final step in the expression of genes. This process is carried out by ribosomes, complexes of protein and specialized RNA molecules. Ribosomes, transfer RNA (tRNA), and other proteins produce a chain of amino acids—the polypeptide—as the end product of translation.
Translation Produces the Building Blocks of...
Translation is the process of synthesizing proteins from the genetic information carried by messenger RNA (mRNA). Following transcription, it constitutes the final step in the expression of genes. This process is carried out by ribosomes, complexes of protein and specialized RNA molecules. Ribosomes, transfer RNA (tRNA), and other proteins produce a chain of amino acids—the polypeptide—as the end product of translation.
Translation Produces the Building Blocks of...
144.0K


