Like hiking? You probably enjoy nature: Persona-grounded Dialog with Commonsense Expansions
This work addresses the challenge of making dialog models more persona-consistent and controllable for applications like chatbots, though it is incremental as it builds on existing methods with enhancements.
The paper tackles the problem of persona-grounded dialog models failing to capture simple implications from persona descriptions, such as inferring that interest in hiking implies love for nature. It proposes expanding persona sentences using commonsense knowledge bases and paraphrasing, along with fine-grained grounding via discrete latent variables, resulting in improved dialog quality and diversity on the PersonaChat dataset.
Existing persona-grounded dialog models often fail to capture simple implications of given persona descriptions, something which humans are able to do seamlessly. For example, state-of-the-art models cannot infer that interest in hiking might imply love for nature or longing for a break. In this paper, we propose to expand available persona sentences using existing commonsense knowledge bases and paraphrasing resources to imbue dialog models with access to an expanded and richer set of persona descriptions. Additionally, we introduce fine-grained grounding on personas by encouraging the model to make a discrete choice among persona sentences while synthesizing a dialog response. Since such a choice is not observed in the data, we model it using a discrete latent random variable and use variational learning to sample from hundreds of persona expansions. Our model outperforms competitive baselines on the PersonaChat dataset in terms of dialog quality and diversity while achieving persona-consistent and controllable dialog generation.