Affect-Driven Dialog Generation
This addresses the need for emotionally appropriate dialog generation in human-computer interaction, but it is incremental as it builds on existing neural methods with added affective controls.
The paper tackles the problem of generating emotional responses in dialog systems by introducing an affect-driven approach that controls affective content using a continuous emotion representation, achieving improved BLEU scores and response diversity.
The majority of current systems for end-to-end dialog generation focus on response quality without an explicit control over the affective content of the responses. In this paper, we present an affect-driven dialog system, which generates emotional responses in a controlled manner using a continuous representation of emotions. The system achieves this by modeling emotions at a word and sequence level using: (1) a vector representation of the desired emotion, (2) an affect regularizer, which penalizes neutral words, and (3) an affect sampling method, which forces the neural network to generate diverse words that are emotionally relevant. During inference, we use a reranking procedure that aims to extract the most emotionally relevant responses using a human-in-the-loop optimization process. We study the performance of our system in terms of both quantitative (BLEU score and response diversity), and qualitative (emotional appropriateness) measures.