arXiv cs.CL· Wei Huang, Keping Bi, Yinqiong Cai, Wei Chen, Jiafeng Guo, Xueqi Cheng·· 2 天前AI 评分24
LLM 生成文本如何影响基于词项的检索模型？
How Do LLM-Generated Texts Impact Term-Based Retrieval Models?
AI 导读
研究发现，LLM 生成文本呈现更平滑的高频 Zipf 斜率、更陡峭的低频斜率和更高的词项特异性，文档级多样性也更强。基于词项的检索模型并不存在固有的来源偏见，而是优先返回词项分布与查询更接近的文档。该工作为理解和应对混合来源内容下基于词项的 IR 系统潜在偏见提供了基础。
正文 · 原文
Abstract:As more content generated by large language models (LLMs) floods into the Internet, information retrieval (IR) systems now face the challenge of distinguishing and handling a blend of human-authored and machine-generated texts. Recent studies suggest that neural retrievers may exhibit a preferential inclination toward LLM-generated content, while classic term-based retrievers like BM25 tend to favor human-written documents. This paper investigates the influence of LLM-generated content on term-based retrieval models, which are valued for their efficiency and robust generalization across domains. Our linguistic analysis reveals that LLM-generated texts exhibit smoother high-frequency and steeper low-frequency Zipf slopes, higher term specificity, and greater document-level diversity. These traits are aligned with LLMs being trained to optimize reader experience through diverse and precise expressions. Our study further explores whether term-based retrieval models demonstrate source bias, concluding that these models prioritize documents whose term distributions closely correspond to those of the queries, rather than displaying an inherent source bias. This work provides a foundation for understanding and addressing potential biases in term-based IR systems managing mixed-source content.
| Subjects: | Information Retrieval (cs.IR); Computation and Language (cs.CL) |
| Cite as: | arXiv:2508.17715 [cs.IR] |
| (or arXiv:2508.17715v2 [cs.IR] for this version) | |
| https://doi.org/10.48550/arXiv.2508.17715 arXiv-issued DOI via DataCite |
Submission history
From: Wei Huang [view email]
[v1]
Mon, 25 Aug 2025 06:43:27 UTC (502 KB)
[v2]
Thu, 15 Oct 2026 05:27:44 UTC (490 KB)
来源：arXiv cs.CL · arxiv.org