arXiv 2603.152532026-03-24cs.CV

HalDec-Bench：图像描述中幻觉检测的基准测试

HalDec-Bench: Benchmarking Hallucination Detector in Image Captioning

Kuniaki Saito, Risa Shinoda, Shohei Tanaka, Tosho Hirasawa, Fumio Okura, Yoshitaka Ushiku

备注：: This work was intended as a replacement of arXiv:2511.20515 and any subsequent updates will appear there

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出HalDec-Bench，用于评估图像描述中幻觉检测的性能，通过多样化模型生成的描述和人工标注揭示了模型在不同难度任务中的表现差异，并发现检测器倾向于认为响应开头的句子正确。

摘要：

图像描述中的幻觉检测（HalDec）通过识别描述中与图像内容不一致的错误来评估视觉-语言模型对图像和文本对齐的能力。除了评估外，有效的幻觉检测对于筛选高质量的图像-描述对以训练VLMs至关重要。然而，由于缺乏全面的基准，VLMs在不同描述模型和幻觉类型上的泛化能力仍不明确。本文介绍了HalDec-Bench，一个旨在以原则性和可解释性评估幻觉检测器的基准。HalDec-Bench包含由多样化VLMs生成的描述，以及指示幻觉存在的手工标注、详细的幻觉类型类别和段落级标签。该基准提供了不同难度级别的任务，并揭示了现有多模态推理或对齐基准中不可见的模型性能差异。我们的分析进一步揭示了两个关键发现。第一，检测器倾向于认为响应开头的句子正确，无论其实际正确性如何。第二，我们的实验表明，通过使用强大的VLMs作为过滤器，结合最近的VLMs作为描述生成器，可以显著减少数据集噪声。我们的项目页面可在https://dahlian00.github.io/HalDec-Bench-Page/上找到。

英文摘要：

Hallucination detection in captions (HalDec) assesses a vision-language model's ability to correctly align image content with text by identifying errors in captions that misrepresent the image. Beyond evaluation, effective hallucination detection is also essential for curating high-quality image-caption pairs used to train VLMs. However, the generalizability of VLMs as hallucination detectors across different captioning models and hallucination types remains unclear due to the lack of a comprehensive benchmark. In this work, we introduce HalDec-Bench, a benchmark designed to evaluate hallucination detectors in a principled and interpretable manner. HalDec-Bench contains captions generated by diverse VLMs together with human annotations indicating the presence of hallucinations, detailed hallucination-type categories, and segment-level labels. The benchmark provides tasks with a wide range of difficulty levels and reveals performance differences across models that are not visible in existing multimodal reasoning or alignment benchmarks. Our analysis further uncovers two key findings. First, detectors tend to recognize sentences appearing at the beginning of a response as correct, regardless of their actual correctness. Second, our experiments suggest that dataset noise can be substantially reduced by using strong VLMs as filters while employing recent VLMs as caption generators. Our project page is available at https://dahlian00.github.io/HalDec-Bench-Page/.

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。