South America
Label Poisoning is All You Need
In a backdoor attack, an adversary injects corrupted data into a model's training dataset in order to gain control over its predictions on images with a specific attacker-defined trigger. A typical corrupted training example requires altering both the image, by applying the trigger, and the label. Models trained on clean images, therefore, were considered safe from backdoor attacks. However, in some common machine learning scenarios, the training labels are provided by potentially malicious third-parties. This includes crowd-sourced annotation and knowledge distillation. We, hence, investigate a fundamental question: can we launch a successful backdoor attack by only corrupting labels?
A Appendix A.1 LangID Details
The complete list may be seen in Table 8. Here are a few general notes about these strings: 1. Based on their recommendations, we did the following: 1. zh, zh_Latn: This resulted in the special filters described below. URLs) the corpora were in languages different from the LangID predictions. This is mainly mis-rendered PDFs and may have practical applications for denoising, or for decoding such garbled PDFs.