基于11类幻觉分类，构建端到端幻觉检测与修正模型。
HAD: HAllucination Detection Language Models Based on a Comprehensive Hallucination Taxonomy
- 提出11类幻觉的完整分类体系，统一检测标准。
- 在9万条合成数据上训练，测试集含2248个样本，性能领先基准。
- 支持多任务幻觉识别与修正，适合实际部署场景。
随着自然语言生成（NLG）模型，特别是大语言模型的广泛应用，其输出的可靠性与准确性引发关注。核心挑战在于幻觉问题——模型生成看似合理但错误的信息。为此，本文提出一个涵盖11类幻觉的综合性分类体系，并构建了集成幻觉检测、片段级定位与修正的端到端模型HAD（https://github.com/pku0xff/HAD）。模型在约9万条合成数据上训练，同时构建了一个包含2248个样本的专用测试集HADTest。在域内与域外测试集上的评估表明，该模型在HaluEval、FactCHD和FaithBench三个基准上均超越现有方法，达到当前最优水平，验证了其鲁棒性与通用性。
原文摘要 · Abstract (English)
The increasing reliance on natural language generation (NLG) models, particularly large language models, has raised concerns about the reliability and accuracy of their outputs. A key challenge is hallucination, where models produce plausible but incorrect information. As a result, hallucination detection has become a critical task. In this work, we introduce a comprehensive hallucination taxonomy with 11 categories across various NLG tasks and propose the HAllucination Detection (HAD) models https://github.com/pku0xff/HAD, which integrate hallucination detection, span-level identification, and correction into a single inference process. Trained on an elaborate synthetic dataset of about 90K samples, our HAD models are versatile and can be applied to various NLG tasks. We also carefully annotate a test set for hallucination detection, called HADTest, which contains 2,248 samples. Evaluations on in-domain and out-of-domain test sets show that our HAD models generally outperform the existing baselines, achieving state-of-the-art results on HaluEval, FactCHD, and FaithBench, confirming their robustness and versatility.
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。