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 2605.00899cs.CVcs.LG
收藏

LatentDiff: 语义数据集比较的扩展方法用于数百万张图像

LatentDiff: Scaling Semantic Dataset Comparison to Millions of Images

School of Electrical Engineering and Computer Science, Oregon State University, Corvallis, Oregon, USA(俄勒冈州立大学电气工程与计算机科学学院);Lawrence Livermore National Laboratory, Livermore, California, USA(劳伦斯利弗莫尔国家实验室);University of California San Diego, California, USA(加州大学圣地亚哥分校)

James Flora, Kowshik Thopalli, Akshay R. Kulkarni, Weng-Keen Wong, Shusen Liu

AI总结:

LatentDiff通过在预训练视觉编码器的潜在空间中结合稀疏自编码器的发散测试与密度比估计,实现语义数据集比较的高效准确,且在极小比例图像差异下仍保持鲁棒性。

摘要:

我们提出了LatentDiff,一种可扩展的语义数据集比较框架,直接在预训练视觉编码器的潜在空间中运行。通过结合基于稀疏自编码器的发散测试与密度比估计,LatentDiff在计算成本仅为基于标签方法的分数的情况下,识别数据集之间的可解释语义差异。我们还引入了Noisy-Diff,一个捕捉现实中的稀疏分布偏移的基准,这些偏移导致现有方法挣扎。实验表明,LatentDiff在极小比例的图像差异(从5%到<1%)下仍保持鲁棒性,同时实现了更高的准确性。

英文摘要:

We present LatentDiff, a scalable framework for semantic dataset comparison that operates directly in the latent space of pretrained vision encoders. By combining sparse autoencoder-based divergence testing with density ratio estimation, LatentDiff identifies interpretable semantic differences between datasets at a fraction of the computational cost of caption-based alternatives. We also introduce Noisy-Diff, a benchmark capturing realistic sparse distribution shifts that cause existing methods to struggle. Experiments demonstrate that LatentDiff achieves superior accuracy while remaining robust to settings where an extremely small fraction of images (from 5% to <1% ) differ semantically.

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。

↑