Leveraging Few-Shot Data Augmentation and Waterfall Prompting for Response Generation

“Task-Oriented Dialogue (TOD) Systems are traditionally designed to facilitate users in achieving specific objectives, such as looking up train times or booking a flight in a dialogue setting. For these tasks, the models are often given access to a database of factual information to complete the task. However, other tasks necessitate not only factual but also subjective insights, which are derived from other users’ opinions. Handling subjective knowledge and using it for generating dialogue responses is the core of the Subjective-Knowledge-based Task- Oriented Dialogue (SK-TOD) (Zhao et al., 2023) challenge. The challenge is set up as conversations between users and artificial assistants, inquiring about and potentially booking hotels or restaurants. The organisers provided dialogue snapshots and a knowledge base with subjective reviews and FAQs related to said hotels and restaurants. The challenge consists of three interlinked subtasks: 1) Knowledge Seeking Turn Detection, where it is determined whether a turn needs knowledge to create an appropriate response; 2) Knowledge Selection, where relevant items are selected from the knowledge base; and 3) Knowledge-Grounded Response Generation, where the dialogue history and the selected knowledge items must be aggregated to generate a concise response for the user.”