arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2406.04175cs.CLcs.AI
收藏

虚构:大型语言模型幻觉的惊人价值

Confabulation: The Surprising Value of Large Language Model Hallucinations

McGill University(麦吉尔大学);Harvard University(哈佛大学)

Peiqi Sui, Eamon Duede, Sophie Wu, Richard Jean So

AI总结:

本文系统性辩护了大型语言模型幻觉作为潜在资源的价值,实证表明幻觉输出比真实输出具有更高的叙事性和语义连贯性,反映其与人类意义建构倾向相似,并与生成连贯叙事文本的积极能力密切相关。

摘要:

本文对大型语言模型(LLM)幻觉或“虚构”作为潜在资源而非绝对负面陷阱进行了系统性辩护。标准观点认为虚构本质上有问题,AI研究应消除这一缺陷。在本文中,我们论证并实证表明,LLM虚构的可测语义特征反映了人类将增强的叙事性作为认知资源进行意义建构和交流的倾向。换言之,它具有潜在价值。具体而言,我们分析了流行的幻觉基准,揭示出与真实输出相比,幻觉输出表现出更高水平的叙事性和语义连贯性。这一发现揭示了我们通常对虚构轻视理解中的张力。它反直觉地表明,LLM虚构的倾向可能与生成连贯叙事文本的积极能力密切相关。

英文摘要:

This paper presents a systematic defense of large language model (LLM) hallucinations or 'confabulations' as a potential resource instead of a categorically negative pitfall. The standard view is that confabulations are inherently problematic and AI research should eliminate this flaw. In this paper, we argue and empirically demonstrate that measurable semantic characteristics of LLM confabulations mirror a human propensity to utilize increased narrativity as a cognitive resource for sense-making and communication. In other words, it has potential value. Specifically, we analyze popular hallucination benchmarks and reveal that hallucinated outputs display increased levels of narrativity and semantic coherence relative to veridical outputs. This finding reveals a tension in our usually dismissive understandings of confabulation. It suggests, counter-intuitively, that the tendency for LLMs to confabulate may be intimately associated with a positive capacity for coherent narrative-text generation.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑