Cue-word Driven Neural Response Generation with a Shrinking Vocabulary
This work addresses the issue of generating diverse and informative responses in dialogue systems, which is incremental as it builds on existing cue-word methods to improve decoding efficiency and diversity.
The paper tackles the problem of generating safe and meaningless responses in open-domain neural dialogue systems by proposing a cue-word driven approach that produces multiple cue-words during decoding to shrink the vocabulary, resulting in significantly outperforming baselines with lower decoding complexity and more efficient convergence to concrete semantics.
Open-domain response generation is the task of generating sensible and informative re-sponses to the source sentence. However, neural models tend to generate safe and mean-ingless responses. While cue-word introducing approaches encourage responses with concrete semantics and have shown tremendous potential, they still fail to explore di-verse responses during decoding. In this paper, we propose a novel but natural approach that can produce multiple cue-words during decoding, and then uses the produced cue-words to drive decoding and shrinks the decoding vocabulary. Thus the neural genera-tion model can explore the full space of responses and discover informative ones with efficiency. Experimental results show that our approach significantly outperforms several strong baseline models with much lower decoding complexity. Especially, our approach can converge to concrete semantics more efficiently during decoding.