AI Slop Is in Your Training Dataset Now. I Tested Three Ways to Spot It.
中文摘要
测试了三种识别训练集 AI 垃圾的方法，发现检测器常误判真人评论，且过滤这些内容会降低模型准确性。
English Summary
Testing three ways to detect AI slop revealed that detectors often misidentify real reviews, and filtering them can decrease model accuracy.
原文节选
My AI detectors flagged many genuine reviews, and filtering them made the sentiment model less accurate. The post AI Slop Is in Your Training Dataset Now. I Tested Three Ways to Spot It. appeared first on Towards Data Science.