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

TL;DR

This paper introduces novel few-shot data augmentation and waterfall prompting techniques to enhance response generation in task-oriented conversational models, leveraging subjective knowledge and multiple GPT models for improved dialogue quality.

Contribution

It presents new methods combining few-shot learning, subjective knowledge augmentation, and waterfall prompting with GPT models for better response generation.

Findings

Improved response relevance and coherence in task-oriented dialogues.

Effective augmentation of subjective knowledge with few-shot learning.

Enhanced performance using waterfall prompting with GPT-3 and ChatGPT.

Abstract

This paper discusses our approaches for task-oriented conversational modelling using subjective knowledge, with a particular emphasis on response generation. Our methodology was shaped by an extensive data analysis that evaluated key factors such as response length, sentiment, and dialogue acts present in the provided dataset. We used few-shot learning to augment the data with newly generated subjective knowledge items and present three approaches for DSTC11: (1) task-specific model exploration, (2) incorporation of the most frequent question into all generated responses, and (3) a waterfall prompting technique using a combination of both GPT-3 and ChatGPT.