OTTers: One-turn Topic Transitions for Open-Domain Dialogue

Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, pages 2492–2504 August 1–6, 2021. ©2021 Association for Computational Linguistics 2492 OTTers: One-turn Topic Transitions for Open-Domain Dialogue Karin Sevegnani, David M. Howcroft, Ioannis Konstas, Verena Rieser The Interaction Lab, MACS Heriot-Watt University Edinburgh, Scotland, UK {karin.sevegnani, i.konstas, v.t.rieser}@hw.ac.uk Abstract Mixed initiative in open-domain dialogue re- quires a system to pro-actively introduce new topics. The one-turn topic transition task ex- plores how a system connects two topics in a cooperative and coherent manner. The goal of the task is to generate a “bridging” utterance connecting the new topic to the topic of the pre- vious conversation turn. We are especially in- terested in commonsense explanations of how a new topic relates to what has been mentioned before. We first collect a new dataset of hu- man one-turn topic transitions, which we call OTTers1. We then explore different strategies used by humans when asked to complete such a task, and notice that the use of a bridging utterance to connect the two topics is the ap- proach used the most. We finally show how existing state-of-the-art text generation models can be adapted to this task and examine the per- formance of these baselines on different splits of the OTTers data. 1 Introduction For a conversation to be truly engaging, we typi- cally assume that both participants take initiative, e.g. by introducing a new topic. We call this a mixed-initiative dialogue. Open-domain systems trained on vast amounts of data (Jiang et al., 2020; Zhang et al., 2020; Gao et al., 2018; Li et al., 2017, 2016; Vinyals and Le, 2015), however, are often purely responsive, make abrupt transitions, or fail to take initiative (see examples in Table 1). In this paper, we consider the case where the system pro-actively introduces a new topic in a conver- sation by providing a commonsense link of how this new topic relates to what was mentioned pre- viously (see Fig.1). We call this transition strategy “bridging”. Humans deploy a range of strategies 1https://github.com/karinseve/OTTers User A Source Topic: I spend a lot of time outside. User B Transition: I like the outdoors as well, espe- cially gardening. It destresses me. Target Topic: I enjoy relaxing and getting flowers. Entity Path: outside - garden - flower User A Source Topic: I like seafood a lot. User B Transition: Since you like seafood, Swedish fish a candy that youis might enjoy? Target Topic: I have no self control when it comes to candy. Entity Path: seafood - Swedish fish - candy User A Source Topic: I think I am getting engaged soon. User B Transition: I have two children from a pre- vious marriage Target Topic: My children are my life. Entity Path: engagement - marriage - child Figure 1: Example topic transitions from OTTers. User A introduces a topic with a short sentence (main con- cept in bold). Then User B responds with a (option- ally …

Content truncated to 500 words