面向数据质量审计的内在自监督方法
Intrinsic Self-Supervision for Data Quality Audits
AI总结：
针对计算机视觉基准数据集存在的跑题图像、近似重复、标签错误等问题，提出名为SelfClean的数据质量审计方法，将数据清理形式化为排序或评分问题，结合上下文感知自监督表示学习与距离指标，在多个常用数据集上性能超SOTA，可识别最高16%的问题并提升评估可靠性。
摘要：
计算机视觉领域的基准数据集通常包含跑题图像、近似重复样本和标签错误，导致模型性能评估结果不准确。本文重新探讨数据清理任务，将其形式化为排序问题（可大幅减少人工检查工作量）或评分问题（可基于分数分布实现自动化决策）。研究发现，将上下文感知自监督表示学习与基于距离的指标特定结合，可有效发现问题且不存在标注偏差。该方法被命名为SelfClean，在检测ImageNet-1k、Food-101N和STL-10等常用图像数据集中的跑题图像、近似重复样本和标签错误方面，性能超越现有最优水平，对合成问题和真实污染均有效。我们将该详细方法应用于多个图像基准，识别出最高达16%的问题，并证实清理后评估可靠性有所提升。官方实现可访问：https://github.com/Digital-Dermatology/SelfClean。
英文摘要：
Benchmark datasets in computer vision often contain off-topic images, near duplicates, and label errors, leading to inaccurate estimates of model performance. In this paper, we revisit the task of data cleaning and formalize it as either a ranking problem, which significantly reduces human inspection effort, or a scoring problem, which allows for automated decisions based on score distributions. We find that a specific combination of context-aware self-supervised representation learning and distance-based indicators is effective in finding issues without annotation biases. This methodology, which we call SelfClean, surpasses state-of-the-art performance in detecting off-topic images, near duplicates, and label errors within widely-used image datasets, such as ImageNet-1k, Food-101N, and STL-10, both for synthetic issues and real contamination. We apply the detailed method to multiple image benchmarks, identify up to 16% of issues, and confirm an improvement in evaluation reliability upon cleaning. The official implementation can be found at: https://github.com/Digital-Dermatology/SelfClean.
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。
