利用损坏的定位数据预训练多模态幻觉检测器
Pre-Training Multimodal Hallucination Detectors with Corrupted Grounding Data
AI总结：
本文提出将多模态幻觉检测建模为序列标注任务，并通过生成损坏的定位数据进行预训练，以提升模型微调时的样本效率和幻觉定位能力。
摘要：
多模态语言模型在输出中可能出现幻觉，这限制了它们的可靠性。自动检测这些错误对于缓解幻觉问题非常重要，但相关探索较少，且现有工作并未对幻觉进行定位，而是将其视为分类任务。在本文中，我们首先将多模态幻觉检测定义为一个序列标注任务，要求模型定位幻觉文本片段，并提出了一个强基线模型。鉴于该任务的人工标注成本高昂，我们提出了一种通过创建损坏的定位数据来提高模型样本效率的方法，并将这些数据用于预训练。我们利用短语定位数据，生成幻觉来替换已定位的片段，从而创建幻觉文本。实验表明，在此数据上进行预训练能够提高微调时的样本效率，并且来自定位数据的学习信号在这些改进中发挥了重要作用。
英文摘要：
Multimodal language models can exhibit hallucinations in their outputs, which limits their reliability. The ability to automatically detect these errors is important for mitigating them, but has been less explored and existing efforts do not localize hallucinations, instead framing this as a classification task. In this work, we first pose multimodal hallucination detection as a sequence labeling task where models must localize hallucinated text spans and present a strong baseline model. Given the high cost of human annotations for this task, we propose an approach to improve the sample efficiency of these models by creating corrupted grounding data, which we use for pre-training. Leveraging phrase grounding data, we generate hallucinations to replace grounded spans and create hallucinated text. Experiments show that pre-training on this data improves sample efficiency when fine-tuning, and that the learning signal from the grounding data plays an important role in these improvements.
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。
