Learning to Generate Prompts for Dialogue Generation through Reinforcement Learning

TL;DR

This paper introduces a reinforcement learning approach combined with multi-task learning to steer dialogue generation models without accessing their parameters, enabling effective control and quick adaptation to new tasks.

Contribution

It proposes a novel method that uses RL and multi-task learning to control dialogue models without parameter access, improving adaptability and generalization.

Findings

Successfully controls SOTA dialogue models without parameter access

Demonstrates rapid adaptation to unseen tasks with fewer training steps

Outperforms baseline models in task generalization

Abstract

Much literature has shown that prompt-based learning is an efficient method to make use of the large pre-trained language model. Recent works also exhibit the possibility of steering a chatbot's output by plugging in an appropriate prompt. Gradient-based methods are often used to perturb the prompts. However, some language models are not even available to the public. In this work, we first explored the combination of prompting and reinforcement learning (RL) to steer models' generation without accessing any of the models' parameters. Second, to reduce the training effort and enhance the generalizability to the unseen task, we apply multi-task learning to make the model learn to generalize to new tasks better. The experiment results show that our proposed method can successfully control several state-of-the-art (SOTA) dialogue models without accessing their parameters. Furthermore, the model demonstrates the strong ability to quickly adapt to an unseen task in fewer steps than the baseline model.