Enoki：高效多層級幻覺偵測

摘要

確保事實性仍然是將大型語言模型部署於高風險場域中的關鍵挑戰。現有的幻覺偵測器通常僅在單一層級運作：主張層級方法提供可解釋的事實單元，而片段層級方法則定位未受文本支持的內容。橋接這兩種觀點的代價高昂，因為重度依賴大型語言模型的流程需要多次分解與驗證呼叫，而模組化系統則需要額外的主張到片段對齊。我們提出Enoki，一個用於多層級幻覺偵測的開放式資訊抽取框架。Enoki抽取以文本為錨定的關係事實，將其與證據進行驗證，並將未受支持的事實映射回幻覺片段。這種共享表徵使得主張層級驗證與片段層級定位得以同時進行，無需分別對齊。Enoki支援基於大型語言模型、基於編碼器及基於規則的抽取模式，透過統一介面在準確度與推論成本之間取得平衡。實驗結果顯示，Enoki在資源消耗較少的情況下仍能與強大的主張層級系統保持競爭力，並在細粒度的片段層級與實體層級定位上達到優異表現。我們亦釋出EnokiQA，一個具備對齊的主張層級驗證與片段層級定位註釋的雙粒度資料集。

English

Ensuring factuality remains a critical challenge for deploying LLMs in high-stakes settings. Existing hallucination detectors usually operate at a single level: claim-level methods provide interpretable factual units, while span-level methods localize unsupported text. Bridging these views is costly, as LLM-heavy pipelines require multiple decomposition and verification calls, and modular systems need additional claim-to-span alignment. We propose Enoki, an Open Information Extraction framework for multi-level hallucination detection. Enoki extracts text-anchored relational facts, verifies them against evidence, and projects unsupported facts back to hallucinated spans. This shared representation enables claim-level verification and span-level localization without requiring separate alignment. Enoki supports LLM-based, encoder-based, and rule-based extraction regimes, balancing accuracy and inference cost through a common interface. Experiments show that Enoki remains competitive with strong claim-level systems while using fewer resources and achieves superior performance on fine-grained span- and entity-level localization. We also release EnokiQA, a dual-granularity dataset with aligned claim-level verification and span-level localization annotations.