The Dialogue Dodecathlon: Open-Domain Knowledge and Image Grounded Conversational Agents

TL;DR

This paper introduces dodecaDialogue, a comprehensive set of 12 tasks to evaluate and improve open-domain conversational agents capable of engaging, reasoning, and perceiving through multi-task learning on diverse datasets.

Contribution

The paper presents a new multi-task benchmark with 12 tasks for open-domain dialogue, demonstrating improved performance and state-of-the-art results across various conversational and perception tasks.

Findings

Multi-tasking improves performance over baseline models.

Large dialogue datasets enhance conversational capabilities.

State-of-the-art results achieved on multiple tasks.

Abstract

We introduce dodecaDialogue: a set of 12 tasks that measures if a conversational agent can communicate engagingly with personality and empathy, ask questions, answer questions by utilizing knowledge resources, discuss topics and situations, and perceive and converse about images. By multi-tasking on such a broad large-scale set of data, we hope to both move towards and measure progress in producing a single unified agent that can perceive, reason and converse with humans in an open-domain setting. We show that such multi-tasking improves over a BERT pre-trained baseline, largely due to multi-tasking with very large dialogue datasets in a similar domain, and that the multi-tasking in general provides gains to both text and image-based tasks using several metrics in both the fine-tune and task transfer settings. We obtain state-of-the-art results on many of the tasks, providing a strong baseline for this challenge.