FutureTOD: Teaching Future Knowledge to Pre-trained Language Model for Task-Oriented Dialogue
This addresses the challenge of adapting pre-trained models for task-oriented dialogues, which is incremental as it builds on existing contrastive frameworks.
The paper tackled the problem of pre-trained language models being less effective for task-oriented dialogues due to linguistic pattern differences, and proposed FutureTOD, a model that distills future knowledge into dialogue representations, achieving improved performance on diverse downstream tasks.
Pre-trained language models based on general text enable huge success in the NLP scenario. But the intrinsical difference of linguistic patterns between general text and task-oriented dialogues makes existing pre-trained language models less useful in practice. Current dialogue pre-training methods rely on a contrastive framework and face the challenges of both selecting true positives and hard negatives. In this paper, we propose a novel dialogue pre-training model, FutureTOD, which distills future knowledge to the representation of the previous dialogue context using a self-training framework. Our intuition is that a good dialogue representation both learns local context information and predicts future information. Extensive experiments on diverse downstream dialogue tasks demonstrate the effectiveness of our model, especially the generalization, robustness, and learning discriminative dialogue representations capabilities.