通过精准注入激活向量，实现情感与谈判行为的高效可控调控。
From Passive to Persuasive: Localized Activation Injection for Empathy and Negotiation
- 定位行为触发的层-令牌位置，进行局部激活注入。
- 在多轮对话中，效果优于全局调控和指令提示。
- 适合需要精细社交行为控制的研究与应用。
复杂社交行为如共情与策略性礼貌，通常被认为难以通过方向分解实现激活调控，而激活调控在粗粒度属性（如情感倾向或毒性）上已证明有效。本文提出STAR（通过归因与表征进行调控），利用归因补丁识别每个行为特征在模型中的因果起源层与令牌位置，随后在这些精确位置注入对比激活向量。在单轮与多轮情绪对话及谈判任务中评估，局部注入始终优于全局调控与指令提示。人工评估确认提升反映真实质量改善，而非表面词汇变化。结果表明，复杂人际行为在大模型激活空间中以局部、近似线性方向编码，行为对齐本质上是定位问题。
原文摘要 · Abstract (English)
Complex social behaviors, such as empathy and strategic politeness, are widely assumed to resist the directional decomposition that makes activation steering effective for coarse attributes like sentiment or toxicity. We present STAR: Steering via Attribution and Representation, which tests this assumption by using attribution patching to identify the layer--token positions where each behavioral trait causally originates, then injecting contrastive activation vectors at precisely those locations. Evaluated on emotional dialogue and negotiation in both single- and multi-turn settings, localized injection consistently outperforms global steering and instruction priming; human evaluation confirms that gains reflect genuine improvements in perceived quality rather than lexical surface change. Our results suggest that complex interpersonal behaviors are encoded as localized, approximately linear directions in LLM activation space, and that behavioral alignment is fundamentally a localization problem.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。