按年被提到
20232023 · 2 次2026
大模型与语言模型›幻觉与语言模型局限·arXiv:2303.08896 · cs.CL

SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Models

Potsawee Manakul, Adian Liusie, Mark J. F. Gales · 2023
开始阅读 ↗PDFalphaXiv 讨论
被提及 · 2 人首次 2023-09-27 · 最近 2024-02-26
2不同的人
2活跃月份
2讨论串
39被引(OpenAlex)
arXiv 摘要(候选层不写导读)

Generative Large Language Models (LLMs) such as GPT-3 are capable of generating highly fluent responses to a wide variety of user prompts. However, LLMs are known to hallucinate facts and make non-factual statements which can undermine trust in their output. Existing fact-checking approaches either require access to the output probability distribution (which may not be available for systems such as ChatGPT) or external databases that are interfaced via separate, often complex, modules. In this work, we propose "SelfCheckGPT", a simple sampling-based approach that can be used to fact-check the responses of black-box models in a zero-resource fashion, i.e. without an external database. SelfCheckGPT leverages the simple idea that if an LLM has knowledge of a given concept, sampled responses are likely to be similar and contain consistent facts. However, for hallucinated facts, stochastically sampled responses are likely to diverge and contradict one another. We investigate this approach by using GPT-3 to generate passages about individuals from the WikiBio dataset, and manually annotate the factuality of the generated passages. We demonstrate that SelfCheckGPT can: i) detect non-factual and factual sentences; and ii) rank passages in terms of factuality. We compare our approach to several baselines and show that our approach has considerably higher AUC-PR scores in sentence-level hallucination detection and higher correlation scores in passage-level factuality assessment compared to grey-box methods.

候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。

HN 书评

推荐在前,同类按时间 · 每篇最多保留 20 条,统计用全部提及
评论og_kalu · 2023-09

“automatically comparing drafts for every single query would be expensive. and that wouldn't eliminate hallucinations just tell you if large details have likely been hallucinated. But it's a method some research has used. https://arxiv.org/abs/2303.08896”

ChatGPT can now see, hear, and speak – openai.com原评论 ↗
评论og_kalu · 2023-09

“https://arxiv.org/abs/2305.18248 "In particular, we find that LMs often hallucinate differing authors of hallucinated references when queried in independent sessions, while consistently identify authors of real references. This suggests that the hallucination may be more a generation issue than inherent to current training techniques or representation." …”

ChatGPT can now see, hear, and speak – openai.com原评论 ↗
评论intended · 2024-02

“https://scholar.google.com/scholar?hl=en&as_sdt=0%2C5&q=llm+hallucination&oq=llm+hallucinatioN https://www.medpagetoday.com/ophthalmology/generalophthalmology/105672#:~:text=The%20mean%20hallucination%20rate%20for,and%2029%25%20with%20version%204.0. - Survey of Hallucination in Natural Language Generation](https://arxiv.org/abs/2202.03629) (Ji et al., 2…”

Hallucination is inevitable: An innate limitation of large language models原评论 ↗