Context is the Key: Backdoor Attacks for In-Context Learning with Vision Transformers

Open in new window