Controllable Mixed-Initiative Dialogue Generation through Prompting
This work addresses the challenge of costly data annotation for conversational agents by enabling more efficient and effective dialogue control, though it is incremental as it builds on existing prompting methods.
The paper tackled the problem of generating controllable mixed-initiative dialogue by replacing fine-tuning with prompting of large language models, resulting in improvements over fine-tuning and ground truth responses in human evaluations and automatic metrics for PersuasionForGood and Emotional Support Conversations tasks.
Mixed-initiative dialogue tasks involve repeated exchanges of information and conversational control. Conversational agents gain control by generating responses that follow particular dialogue intents or strategies, prescribed by a policy planner. The standard approach has been fine-tuning pre-trained language models to perform generation conditioned on these intents. However, these supervised generation models are limited by the cost and quality of data annotation. We instead prompt large language models as a drop-in replacement to fine-tuning on conditional generation. We formalize prompt construction for controllable mixed-initiative dialogue. Our findings show improvements over fine-tuning and ground truth responses according to human evaluation and automatic metrics for two tasks: PersuasionForGood and Emotional Support Conversations.