AttentionSwitch: Jailbreak Defense for LLMs via Joint Learning of Sparse Attention-Head Selection and Scaling

arXiv:2611.08752 · cs.AI, cs.CL, cs.CR · Submitted 2026-11-05 · Read on arXiv

cs.AI, cs.CL, cs.CR

Submitted: 2026-11-05

Updated: 2026-11-13

Terminology

Sources

Related papers