不要失去焦点:通过键正交投影进行激活引导
Don't Lose Focus: Activation Steering via Key-Orthogonal Projections
AI总结:
本文提出SKOP方法,通过约束有害的注意力重路由,减少激活引导对推理和检索性能的负面影响,实现最佳的引导-实用性平衡。
摘要:
激活引导通过干预内部表示来控制LLM行为,但往往会导致推理和检索性能下降。我们提出SKOP,一种通过键正交投影进行激活引导的方法,通过保留模型依赖的少量焦点标记的注意力模式,同时允许在次要尾部标记上重新分配注意力,从而约束有害的注意力重路由。在多个引导基准测试中,SKOP实现了最佳的引导-实用性平衡,减少实用性下降幅度达5-7倍,同时保留超过95%的原始引导效果。我们的结果进一步表明,在长上下文检索设置中,当原始引导方法无效时,SKOP可通过避免注意力重路由保持稳健性能。
英文摘要:
Activation steering controls LLM behaviour towards target behaviour by intervening in internal representations, yet it often degrades reasoning and retrieval performance. We argue that a primary cause of this trade-off is attention rerouting: steering vectors alter query-key matching, shifting attention away from contextually important tokens toward less informative ones. To address this, we propose Steering via Key-Orthogonal Projections (SKOP), a steering method that constrains harmful attention rerouting without eliminating steering efficacy. SKOP achieves this by preserving attention patterns on a small set of focus tokens the model relies on for reasoning and retrieval, while allowing redistribution among less critical tail tokens. Across multiple steering benchmarks, we show that SKOP achieves the best joint steering-utility trade-off, reducing utility degradation by 5-7x while retaining over 95% of vanilla steering efficacy. Our results further suggest that, in long-context retrieval settings where vanilla steering approaches are ineffective, SKOP can maintain robust performance by avoiding attention rerouting.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。