Generating Dialogue Responses from a Semantic Latent Space
This addresses the issue of generating diverse and meaningful dialogue responses for conversational AI systems, representing an incremental improvement over existing methods.
The paper tackled the problem of generic and uninformative responses in open-domain dialogue generation by proposing a model that learns prompt-response relationships as regression on a latent space, resulting in responses that are both relevant and informative according to human evaluation.
Existing open-domain dialogue generation models are usually trained to mimic the gold response in the training set using cross-entropy loss on the vocabulary. However, a good response does not need to resemble the gold response, since there are multiple possible responses to a given prompt. In this work, we hypothesize that the current models are unable to integrate information from multiple semantically similar valid responses of a prompt, resulting in the generation of generic and uninformative responses. To address this issue, we propose an alternative to the end-to-end classification on vocabulary. We learn the pair relationship between the prompts and responses as a regression task on a latent space instead. In our novel dialog generation model, the representations of semantically related sentences are close to each other on the latent space. Human evaluation showed that learning the task on a continuous space can generate responses that are both relevant and informative.