Open Domain Dialogue Generation with Latent Images

TL;DR

This paper introduces a novel method for open domain dialogue generation that leverages latent images to ground responses, enabling effective augmentation and content enrichment in both image-grounded and text-based conversations.

Contribution

It proposes a conditional variational auto-encoding framework to recover latent images from textual dialogues and integrate them into response generation.

Findings

Latent images improve response relevance in text-based conversations.

Textual dialogues can augment image-grounded dialogues, especially in low-resource settings.

Latent images enrich response content while maintaining contextual relevance.

Abstract

We consider grounding open domain dialogues with images. Existing work assumes that both an image and a textual context are available, but image-grounded dialogues by nature are more difficult to obtain than textual dialogues. Thus, we propose learning a response generation model with both image-grounded dialogues and textual dialogues by assuming that the visual scene information at the time of a conversation can be represented by an image, and trying to recover the latent images of the textual dialogues through text-to-image generation techniques. The likelihood of the two types of dialogues is then formulated by a response generator and an image reconstructor that are learned within a conditional variational auto-encoding framework. Empirical studies are conducted in both image-grounded conversation and text-based conversation. In the first scenario, image-grounded dialogues, especially under a low-resource setting, can be effectively augmented by textual dialogues with latent images; while in the second scenario, latent images can enrich the content of responses and at the same time keep them relevant to contexts.