2023/06/17 by Weihao Zeng, Zeng, Weihao, Keqing He +11
Computer Science · Psychology · #Artificial intelligence #Cognitive science #Computation and Language (cs.CL) #Computer science #Discriminative model #FOS: Computer and information sciences #Generalization #Intuition #Language model #Natural Language Processing Techniques #Natural language processing #Psychology #Robustness (evolution) #Speech and dialogue systems #Task (project management) #Topic Modeling
paper · pdf · doi:10.48550/arxiv.2306.10315
published in arXiv (Cornell University) (Cornell University)
openalex publication_date 2023/06/17 · openalex created_date 2023/06/22 · openalex updated_date 2026/07/28
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.