Learning to Generate Prompts for Dialogue Generation through Reinforcement Learning
This work addresses the challenge of controlling black-box dialogue models for developers and researchers, though it is incremental as it builds on existing prompting and RL techniques.
The paper tackled the problem of steering dialogue models' outputs without accessing their parameters by combining prompt-based learning with reinforcement learning and multi-task learning, achieving successful control of several SOTA models and faster adaptation to unseen tasks with fewer steps than baselines.
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.