A Simple And Efficient Multi-task Learning Approach For Conditioned Dialogue Generation · The Large Language Model Bible Contribute to LLM-Bible

A Simple And Efficient Multi-task Learning Approach For Conditioned Dialogue Generation

Zeng Yan, Nie Jian-yun. Arxiv 2020

[Paper]    
Model Architecture Pretraining Methods RAG Transformer

Conditioned dialogue generation suffers from the scarcity of labeled responses. In this work, we exploit labeled non-dialogue text data related to the condition, which are much easier to collect. We propose a multi-task learning approach to leverage both labeled dialogue and text data. The 3 tasks jointly optimize the same pre-trained Transformer – conditioned dialogue generation task on the labeled dialogue data, conditioned language encoding task and conditioned language generation task on the labeled text data. Experimental results show that our approach outperforms the state-of-the-art models by leveraging the labeled texts, and it also obtains larger improvement in performance comparing to the previous methods to leverage text data.

Similar Work