Aiming to Know You Better Perhaps Makes Me a More Engaging Dialogue Partner
This work addresses the challenge of improving engagement in dialogue systems for users, though it appears incremental as it builds on existing attempts to define motivations for agents.
The paper tackled the problem of creating engaging chit-chat dialogue agents by focusing on discovering information about the interlocutor, proposing a quantitative metric and algorithm, and validating it with human evaluation where the system outperformed baselines and showed correlation with engagingness judgments.
There have been several attempts to define a plausible motivation for a chit-chat dialogue agent that can lead to engaging conversations. In this work, we explore a new direction where the agent specifically focuses on discovering information about its interlocutor. We formalize this approach by defining a quantitative metric. We propose an algorithm for the agent to maximize it. We validate the idea with human evaluation where our system outperforms various baselines. We demonstrate that the metric indeed correlates with the human judgments of engagingness.