AI & ML
impact 16
Backdoor Attacks on Prompt-Driven Video Segmentation Foundation Models
Backdoor Attacks on Prompt-Driven Video Segmentation Foundation Models arXiv:2512.22046v2 Announce Type: replace-cross Abstract: Prompt-driven Video Segmentation Foundation Models (VSFMs), such as SAM2, are increasingly…
Why it matters
This adds a new dimension to the promptdriven conversation. Practitioners should assess exposure to video changes.