THRONE：一个面向大型视觉语言模型自由生成式的基于对象的幻觉基准
THRONE: An Object-based Hallucination Benchmark for the Free-form Generations of Large Vision-Language Models
AI总结：
针对大型视觉语言模型自由生成中的“第一类幻觉”问题，提出基于对象的自动化评估框架THRONE，利用公开语言模型量化幻觉，并发现其与“第二类幻觉”负相关，同时提供数据增强基线以减少两类幻觉。
摘要：
缓解大型视觉语言模型（LVLMs）中的幻觉仍然是一个开放问题。现有的基准并未解决开放式自由格式回答中的幻觉问题，我们将此称为“第一类幻觉”。相反，它们专注于针对特定问题格式的幻觉——通常是关于特定对象或属性的多项选择回答——我们将此称为“第二类幻觉”。此外，此类基准通常需要调用可能随时变动的外部API模型。在实践中，我们观察到第二类幻觉的减少并未导致第一类幻觉的减少，相反，这两种形式的幻觉通常是负相关的。为了解决这一问题，我们提出了THRONE，一种新颖的基于对象的自动化框架，用于定量评估LVLM自由格式输出中的第一类幻觉。我们使用公开的语言模型（LMs）来识别LVLM回答中的幻觉，并计算信息丰富的指标。通过使用公开数据集评估大量近期的LVLM，我们表明现有指标的改善并未导致第一类幻觉的减少，且用于测量第一类幻觉的现有基准是不完整的。最后，我们提供了一种简单有效的数据增强方法来减少第一类和第二类幻觉，以此作为一个强有力的基线。代码现已发布于 https://github.com/amazon-science/THRONE 。
英文摘要：
Mitigating hallucinations in large vision-language models (LVLMs) remains an open problem. Recent benchmarks do not address hallucinations in open-ended free-form responses, which we term "Type I hallucinations". Instead, they focus on hallucinations responding to very specific question formats -- typically a multiple-choice response regarding a particular object or attribute -- which we term "Type II hallucinations". Additionally, such benchmarks often require external API calls to models which are subject to change. In practice, we observe that a reduction in Type II hallucinations does not lead to a reduction in Type I hallucinations but rather that the two forms of hallucinations are often anti-correlated. To address this, we propose THRONE, a novel object-based automatic framework for quantitatively evaluating Type I hallucinations in LVLM free-form outputs. We use public language models (LMs) to identify hallucinations in LVLM responses and compute informative metrics. By evaluating a large selection of recent LVLMs using public datasets, we show that an improvement in existing metrics do not lead to a reduction in Type I hallucinations, and that established benchmarks for measuring Type I hallucinations are incomplete. Finally, we provide a simple and effective data augmentation method to reduce Type I and Type II hallucinations as a strong baseline. Code is now available at https://github.com/amazon-science/THRONE .
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。
