An Affect-Rich Neural Conversational Model with Biased Attention and Weighted Cross-Entropy Loss
This work addresses the challenge of embedding affect into human-machine conversations, which is important for improving AI communication, but it is incremental as it builds on existing Seq2Seq models with specific enhancements.
The authors tackled the problem of generating affect-rich responses in open-domain neural conversational models by proposing an end-to-end model that embeds words with VAD affective notations, uses a biased affective attention mechanism, and trains with an affect-incorporated objective function. The model outperformed the state-of-the-art baseline in both perplexity and human evaluations for producing natural and affect-rich responses.
Affect conveys important implicit information in human communication. Having the capability to correctly express affect during human-machine conversations is one of the major milestones in artificial intelligence. In recent years, extensive research on open-domain neural conversational models has been conducted. However, embedding affect into such models is still under explored. In this paper, we propose an end-to-end affect-rich open-domain neural conversational model that produces responses not only appropriate in syntax and semantics, but also with rich affect. Our model extends the Seq2Seq model and adopts VAD (Valence, Arousal and Dominance) affective notations to embed each word with affects. In addition, our model considers the effect of negators and intensifiers via a novel affective attention mechanism, which biases attention towards affect-rich words in input sentences. Lastly, we train our model with an affect-incorporated objective function to encourage the generation of affect-rich words in the output responses. Evaluations based on both perplexity and human evaluations show that our model outperforms the state-of-the-art baseline model of comparable size in producing natural and affect-rich responses.