From Rational Answers to Emotional Resonance: The Role of Controllable Emotion Generation in Language Models

TL;DR

This paper introduces a novel method for controllably generating emotions in large language models using Emotion Vectors, enabling more natural and emotionally resonant human-AI interactions without retraining the models.

Contribution

It proposes a new EV steering framework that allows fine-grained emotional modulation in LLMs through internal activation shifts, without additional training or architecture changes.

Findings

Achieves consistent emotional alignment across multiple LLMs

Maintains semantic fidelity and linguistic fluency during emotion control

Outperforms existing prompt-based and fine-tuning baselines in flexibility and generalizability

Abstract

Purpose: Emotion is a fundamental component of human communication, shaping understanding, trust, and engagement across domains such as education, healthcare, and mental health. While large language models (LLMs) exhibit strong reasoning and knowledge generation capabilities, they still struggle to express emotions in a consistent, controllable, and contextually appropriate manner. This limitation restricts their potential for authentic human-AI interaction. Methods: We propose a controllable emotion generation framework based on Emotion Vectors (EVs) - latent representations derived from internal activation shifts between neutral and emotion-conditioned responses. By injecting these vectors into the hidden states of pretrained LLMs during inference, our method enables fine-grained, continuous modulation of emotional tone without any additional training or architectural modification. We further provide theoretical analysis proving that EV steering enhances emotional expressivity while maintaining semantic fidelity and linguistic fluency. Results: Extensive experiments across multiple LLM families show that the proposed approach achieves consistent emotional alignment, stable topic adherence, and controllable affect intensity. Compared with existing prompt-based and fine-tuning-based baselines, our method demonstrates superior flexibility and generalizability. Conclusion: Emotion Vector (EV) steering provides an efficient and interpretable means of bridging rational reasoning and affective understanding in large language models, offering a promising direction for building emotionally resonant AI systems capable of more natural human-machine interaction.