大语言模型幻觉的几何分类
A Geometric Taxonomy of Hallucinations in LLMs
AI总结：
本文提出基于嵌入空间几何的幻觉检测方法，通过三种操作类型区分可检测与不可检测的幻觉，验证了在不同领域数据集上的有效性。
摘要：
部署中的语言模型幻觉可能对医疗、法律和金融服务等领域的下游决策产生实际影响。在生产环境中，检测必须基于部署系统可见的内容：查询、响应和通常的源文档。白盒访问模型内部和多样本查询通常无法通过第三方API获得。在这一设定下，即黑盒、单次通过、仅问题/回答可用的情况下，主导的基线是NLI，其在失败时仅返回一个值而无诊断。我们主张直接操作嵌入空间的几何结构，提供检测方法，其成功与失败可解释为对比句编码训练的结构性质。贡献是：给定一个操作性动机的分类，几何预测哪些类型的幻觉可检测，哪些不可检测，并且预测成立。我们提出三种操作类型，按响应嵌入与基于单位超球面的 grounded 响应可信区域的关系组织，并从对齐目标推导出每种类型的预测：（1）查询附近的不忠实可通过角度比检测；（2）在可信区域外的编造产生方向性特征，其在专家标注的错误上优于NLI；（3）事实性错误共享词汇和框架与正确答案不可通过角度几何区分。为了在与部署相似的内容上验证，我们构建了一个包含212对人工编造数据集，涵盖九个领域，使用诱发编造方法。
英文摘要：
Hallucinations in deployed language models can have real consequences for downstream decisions in domains such as healthcare, legal, and financial services. In production, detection has to run on what the deployed system can see: the query, the response, and often a source document. White-box access to model internals and multi-sample querying are not generally available behind a third-party API. Within this setting - black-box, single-pass, only question/answer available - the dominant baseline is NLI, which returns a value but no diagnosis when it fails. We argue that operating directly on the geometry of the embedding space provides detection methods whose successes and failures are interpretable as structural properties of contrastive sentence-encoder training \citep{wang2020understanding}. The contribution is: given an operationally-motivated taxonomy, geometry predicts which types of hallucination are detectable and which are not - and the predictions hold. We propose three operational types organized by the relation of the response embedding to the plausibility region of grounded responses on the unit hypersphere, and derive from the alignment objective a prediction for each: (1)query-proximate unfaithfulness is detectable by an angular ratio; (2)confabulation outside the plausibility region produces a directional signature that outperforms NLI on expert-annotated error; (3)factual errors sharing vocabulary and frame with correct answers are not separable by angular geometry. To validate on content resembling deployment, we built a 212-pair human-confabulated dataset across nine domains using provoked confabulation.
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。
