BEAF：通过观察前后变化评估视觉语言模型中的幻觉
BEAF: Observing BEfore-AFter Changes to Evaluate Hallucination in Vision-language Models
AI总结：
本文提出BEAF基准，通过图像编辑改变场景并设计四项新指标，系统评估视觉语言模型对场景变化的感知能力，以揭示并量化幻觉现象。
摘要：
视觉语言模型（VLMs）通过视觉编码器和大语言模型（LLM）的结合来感知世界。视觉编码器在大规模视觉-文本数据集上预训练，为视觉数据提供零样本泛化能力，而LLM赋予VLMs高推理能力。这使得VLMs无需微调即可在广泛基准上取得高性能，展现出零样本或少样本能力。然而，近期研究表明VLMs易受幻觉影响。这种不良行为降低了可靠性和可信度，使用户无法完全信任VLMs的输出。为增强可信度并更好地应对VLMs的幻觉，我们构建了一个新的评估数据集，称为前后幻觉数据集（BEAF），并引入了新指标：真正理解（TU）、忽视（IG）、固执（SB）和犹豫（ID）。与以往仅关注构建问题和答案的工作不同，我们基准的关键思想是通过图像编辑模型操控视觉场景信息，并基于场景变化设计指标。这使我们能够通过观察VLMs感知变化的能力，清晰评估其是否正确理解给定场景。我们还借助视觉和文本的双轴视图可视化图像级对象关系。使用我们的数据集评估VLMs时，我们观察到这些指标揭示了以往未报告的VLM幻觉的不同方面。项目页面：\url{https://beafbench.github.io/}
英文摘要：
Vision language models (VLMs) perceive the world through a combination of a visual encoder and a large language model (LLM). The visual encoder, pre-trained on large-scale vision-text datasets, provides zero-shot generalization to visual data, and the LLM endows its high reasoning ability to VLMs. It leads VLMs to achieve high performance on wide benchmarks without fine-tuning, exhibiting zero or few-shot capability. However, recent studies show that VLMs are vulnerable to hallucination. This undesirable behavior degrades reliability and credibility, thereby making users unable to fully trust the output from VLMs. To enhance trustworthiness and better tackle the hallucination of VLMs, we curate a new evaluation dataset, called the BEfore-AFter hallucination dataset (BEAF), and introduce new metrics: True Understanding (TU), IGnorance (IG), StuBbornness (SB), and InDecision (ID). Unlike prior works that focus only on constructing questions and answers, the key idea of our benchmark is to manipulate visual scene information by image editing models and to design the metrics based on scene changes. This allows us to clearly assess whether VLMs correctly understand a given scene by observing the ability to perceive changes. We also visualize image-wise object relationship by virtue of our two-axis view: vision and text. Upon evaluating VLMs with our dataset, we observed that our metrics reveal different aspects of VLM hallucination that have not been reported before. Project page: \url{https://beafbench.github.io/}
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。
