AI RESEARCH

Delayed Backdoor Attacks: Exploring the Temporal Dimension as a New Attack Surface in Pre-Trained Models

arXiv CS.AI

ArXi:2603.11949v1 Announce Type: cross Backdoor attacks against pre-trained models (PTMs) have traditionally operated under an ``immediacy assumption,'' where malicious behavior manifests instantly upon trigger occurrence. This work revisits and challenges this paradigm by