利用未来上下文提升幻觉检测准确率
Enhancing Hallucination Detection via Future Context
- 通过采样未来上下文捕捉幻觉的持续性特征
- 在多种方法上实现检测性能显著提升
- 适合关注黑箱生成模型可信度的研究者
大型语言模型（LLMs）被广泛用于在线平台生成看似合理的文本，但其生成过程不透明。随着用户越来越多地接触此类黑箱输出，幻觉检测成为关键挑战。为此，本文针对黑箱生成器设计了一种幻觉检测框架。基于幻觉一旦产生便倾向于持续存在的观察，我们采样未来上下文，这些上下文为幻觉检测提供了重要线索，并可有效融入多种基于采样的方法中。我们在多种方法上广泛验证了该采样策略带来的性能提升。
原文摘要 · Abstract (English)
Large Language Models (LLMs) are widely used to generate plausible text on online platforms, without revealing the generation process. As users increasingly encounter such black-box outputs, detecting hallucinations has become a critical challenge. To address this challenge, we focus on developing a hallucination detection framework for black-box generators. Motivated by the observation that hallucinations, once introduced, tend to persist, we sample future contexts. The sampled future contexts provide valuable clues for hallucination detection and can be effectively integrated with various sampling-based methods. We extensively demonstrate performance improvements across multiple methods using our proposed sampling approach.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。