发表机构
Sharda University; University of Maryland, Baltimore County（沙尔达大学; 马里兰大学巴尔的摩县分校）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本调查基于LLM生命周期对幻觉进行分类，探讨其成因、检测、缓解与预防，分析基准数据适用性，为相关人员提供标准化框架以构建更可靠的LLM。
AI 中文摘要
大语言模型（LLM）中幻觉的生命周期这一概念，为构建高风险环境（包括医疗、法律及科学研究领域）中LLM控制与可靠性的可靠框架提供了支撑。尽管以往调查主要聚焦于检测或缓解环节，但本调查基于LLM生命周期对其幻觉现象、原因、检测、缓解及预防展开了概述。本调查提出了LLM生命周期中幻觉的三重分类：数据相关、训练相关及推理相关，与LLM的开发周期相一致。针对每个阶段，本调查均讨论了幻觉的成因、检测方式，以及在特定缓解或预防干预措施下的解决途径。此外，本调查还利用多项参数对可用基准数据进行了探讨，以确定其在识别、限制及管理幻觉方面的适用性。本调查为研究人员与从业者提供了标准化框架，使其能够系统地理解、诊断并应对幻觉，从而构建更安全、更可靠的LLM。
英文摘要
The lifecycle of hallucination in LLMs is a concept that enables building solid frameworks on the control and reliability of LLMs in high-stakes environments, including health, legal, and scientific research. Although previous surveys have primarily focused on detection or mitigation, this survey provides a lifecycle-based overview of the hallucinations in the LLMs, their cause, detection, mitigation, and prevention.We propose a three-fold categorization of hallucinations across the LLM lifecycle: data-related, training-related, and inference-related, which is consistent with the lifecycle of the development of the LLM. Each of these stages is discussed regarding the cause of hallucinations, their detection, and the ways they can be addressed under specific mitigation or prevention interventions. In addition, we discuss the available benchmark data using a number of parameters so as to establish their suitability in identifying, restricting and managing hallucinations. The survey provides researchers and practitioners with a standardized framework to understand, diagnose, and cure hallucinations in a systematic system to present actionable data to build safer and more reliable LLMs.
Journal refLamba, N., Tiwari, S. & Gaur, M. Hallucinations in LLMs: A Lifecycle-Based Survey of Causes, Detection, Mitigation, and Prevention. International Journal of Data Science and Analytics 22, 232 (2026)
DOI：10.1007/s41060-026-01214-6
