AIIC AI Intelligence Centre

SOURCE-LINKED INTELLIGENCE

ShallowDefend: Harnessing Shallow Safety Alignment for Multi-Turn Defender

arXiv · AI, language, vision and robotics · article · Oct 29, 2026 · UTC

Safety alignment has enabled large language models to reliably refuse harmful requests directly, yet the same request distributed across a multi-turn conversation still succeeds far too often. This brittleness has a mechanistic root: Alignment is shallow, with safety behavior concentrated in a model's first few output tokens, and existing defenses inherit the weakness, supervising the surface form of the response without ever giving the model an explicit safety decision to learn. We take the opposite view of shallowness: if a short span determines the trajectory of an entire response, that spa

Read original source ↗ Open in workspace

recordType
paper
region
Global

Evidence & attribution

First collected: 2026-10-31T14:12:28.156Z. This is not the publication date.