emotion-control

PragAlign introduces a feedback-guided framework for controlled synthetic dialogue generation using an LLM-based evaluator to improve alignment with intent, emotion, coherence, and fluency, achieving 99.50% acceptance compared to 72.25% for one-shot generation.

This paper proposes EmoVec, a lightweight framework for controllable affective generation in large language models via latent vector steering, enabling continuous control over emotional intensity without model weight updates.

This paper proposes VA-DPO, a method for controllable emotion generation in language models using continuous valence-arousal dimensions, which improves over prompting techniques without degrading model performance.

NeuTTS-2E is an open-source on-device TTS model that supports seven controllable emotions.