大型语言模型幻觉检测
背景 1.1 幻觉检测的定义与范围 大型语言模型（LLMs）中的幻觉检测 是指系统性地识别由LLMs生成的事实错误或无意义输出的任务，而无需依赖外部证据 [Li et al., 2024; Zhang et al., 2024]。这项任务对于确保LLM生成内容的可靠性和可信度至关重要，特别是在医疗、法律咨询...
Posts by Year
2025
Llm Hallucination Detection Method Survey
Methodology for Hallucination in Large Language Models
Llm Hallucination Detection Dataset Survey
Benchmarking the Benchmarks: A Systematic Survey of Datasets for Hallucination Detection in Large Language Models 关键词： Large Language Models, Hallucination ...