发表机构
Gaoling School of Artificial Intelligence, Renmin University of China; Tencent Jarvis Lab（中国人民大学高瓴人工智能学院; 腾讯Jarvis实验室）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对大语言模型对齐后情感响应平淡的问题，提出轻量框架EmoVec，通过调控潜在向量实现可控情感生成，实验验证其能提升情感显著性且保留语义等特性。
AI 中文摘要
大语言模型（LLMs）在对齐后常产生情感平淡的响应，限制了其在情感敏感型应用中的效能。本文提出EmoVec，一种通过潜在向量调控实现可控情感生成的轻量框架。EmoVec利用对比激活加法从配对的中性响应与情感条件响应中提取特定情感方向，再通过任务特定去偏与主子空间移除对其进行优化。推理阶段，这些向量以静态或场景自适应缩放方式注入最终残差流，无需更新模型权重即可实现对情感强度的连续控制。在三个LLMs和八种情感上开展的实验表明，EmoVec在提升情感显著性的同时，能较好保留语义内容、流畅度与连贯性。消融研究与人工评估进一步验证了向量纯化与自适应缩放的有效性，确立EmoVec为部署LLMs中用于情感控制的实用推理时方法。
英文摘要
Large Language Models (LLMs) often produce emotionally flattened responses after alignment, limiting their effectiveness in affect-sensitive applications. In this paper, we propose EmoVec, a lightweight framework for controllable affective generation via latent vector steering. EmoVec extracts emotion-specific directions from paired neutral and emotion-conditioned responses using contrastive activation addition, and further refines them through task-specific debiasing and principal subspace removal. During inference, these vectors are injected into the final residual stream with static or scenario-adaptive scaling, enabling continuous control over emotional intensity without updating model weights. Experiments across three LLMs and eight emotions show that EmoVec consistently improves emotional salience while largely preserving semantic content, fluency, and coherence. Ablation studies and human evaluation further confirm the effectiveness of vector purification and adaptive scaling, establishing EmoVec as a practical inference-time method for affective control in deployed LLMs.
