IHDec: Divergence-Steered Contrastive Decoding for Securing Multi-Turn Instruction Hierarchies

Excerpt

arXiv:2606.29960v2 Announce Type: replace Abstract: Large Language Models (LLMs) often fail to maintain instruction hierarchies (IH) when processing multi-source inputs with varying role-level priorities, paradoxically adhering to lower-priority directives during conflicts. While existing defenses mitigate this issue, they are largely restricted to single-turn scenarios and require expensive fine-tuning. In this paper, we formalize this failure mode in multi-turn contexts via a Jensen-Shannon Di