方法对比

	幻觉检测	问答 (QA)
领域	文本挖掘	文本挖掘
方法族	Process / pipeline	Process / pipeline
起源年份	2020 (faithfulness framing); 2023 (SelfCheckGPT)	—
提出者	Established as a formal task by Maynez et al. (2020); SelfCheckGPT zero-resource variant by Manakul et al. (2023)	—
类型	NLP evaluation / quality-assurance pipeline	NLP text-comprehension task
开创性文献	Maynez, J., Narayan, S., Bohnet, B., & McDonald, R. (2020). On Faithfulness and Factuality in Abstractive Summarization. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (ACL), 1906-1919.	Rajpurkar, P. et al. (2016). SQuAD: 100,000+ Questions for Machine Comprehension of Text. EMNLP.
别名	factual consistency checking, faithfulness evaluation, LLM output verification, Hallüsinasyon Tespiti (Factual Consistency)	QA, machine reading comprehension, Soru Cevaplama (Question Answering)
相关	5	4
摘要	Hallucination detection is a natural-language-processing pipeline that measures whether the output of a language model is consistent with a reference source document or with verifiable facts. Formalised as a faithfulness evaluation task by Maynez et al. (2020) and extended to a zero-resource black-box setting by Manakul et al. (2023) with SelfCheckGPT, the approach is used to flag unreliable LLM outputs in high-stakes domains such as medicine, law, and journalism.	Question answering is a natural-language-processing task that automatically answers natural-language questions grounded in a given context passage, using either extractive or generative approaches. The task was crystallised by the SQuAD benchmark of Rajpurkar et al. (2016), and later models such as XLNet (Yang et al., 2019) pushed reading-comprehension accuracy higher.