arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2602.09475cs.CVcs.AIcs.LG
收藏

ArtifactLens:数百个标签足以让VLMs检测到瑕疵

ArtifactLens: Hundreds of Labels Are Enough for Artifact Detection with VLMs

Stanford University, Stanford, CA, USA(斯坦福大学);Snap Inc., Santa Monica, CA, USA(Snap公司)

James Burgess, Rameen Abdal, Dan Stoddart, Sergey Tulyakov, Serena Yeung-Levy, Kuan-Chieh Jackson Wang

AI总结:

ArtifactLens通过少量标记数据利用预训练VLMs的潜力,实现高效瑕疵检测,适用于多种瑕疵类型和AIGC检测任务。

摘要:

现代图像生成器生成出令人惊叹的逼真图像,其中只有像扭曲的手或变形的物体这样的瑕疵揭示出其合成的起源。检测这些瑕疵至关重要:没有检测,我们无法对生成器进行基准测试,也无法训练奖励模型来改进它们。当前的检测器在数万张标记图像上微调VLMs,但每次生成器演变或新瑕疵类型出现时,这种做法都显得昂贵。我们显示,预训练的VLMs已经编码了检测瑕疵所需的知识——只要有了正确的支架,这种能力可以通过仅几百个标记示例每种瑕疵类别来解锁。我们的系统ArtifactLens在五个人类瑕疵基准测试中实现了最先进的性能(首次在多个数据集上进行评估),同时需要数量级更少的标记数据。支架由一个多组件架构组成,包括上下文学习和文本指令优化,并对每个部分都有新的改进。我们的方法可以推广到其他瑕疵类型——物体形态、动物解剖学和实体互动——以及AIGC检测这一不同的任务。

英文摘要:

Modern image generators produce strikingly realistic images, where only artifacts like distorted hands or warped objects reveal their synthetic origin. Detecting these artifacts is essential: without detection, we cannot benchmark generators or train reward models to improve them. Current detectors fine-tune VLMs on tens of thousands of labeled images, but this is expensive to repeat whenever generators evolve or new artifact types emerge. We show that pretrained VLMs already encode the knowledge needed to detect artifacts - with the right scaffolding, this capability can be unlocked using only a few hundred labeled examples per artifact category. Our system, ArtifactLens, achieves state-of-the-art on five human artifact benchmarks (the first evaluation across multiple datasets) while requiring orders of magnitude less labeled data. The scaffolding consists of a multi-component architecture with in-context learning and text instruction optimization, with novel improvements to each. Our methods generalize to other artifact types - object morphology, animal anatomy, and entity interactions - and to the distinct task of AIGC detection.

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。

↑