Query Enhanced Knowledge-Intensive Conversation via Unsupervised Joint Modeling
This work addresses the challenge of knowledge-intensive conversations for AI systems, offering an incremental improvement by enabling unsupervised learning without extra annotations.
The paper tackles the problem of generating responses in knowledge-intensive conversations without requiring annotated queries or knowledge sources, by proposing an unsupervised joint modeling approach called QKConv. The method outperforms all unsupervised baselines across three datasets and achieves competitive performance compared to supervised methods.
In this paper, we propose an unsupervised query enhanced approach for knowledge-intensive conversations, namely QKConv. There are three modules in QKConv: a query generator, an off-the-shelf knowledge selector, and a response generator. QKConv is optimized through joint training, which produces the response by exploring multiple candidate queries and leveraging corresponding selected knowledge. The joint training solely relies on the dialogue context and target response, getting exempt from extra query annotations or knowledge provenances. To evaluate the effectiveness of the proposed QKConv, we conduct experiments on three representative knowledge-intensive conversation datasets: conversational question-answering, task-oriented dialogue, and knowledge-grounded conversation. Experimental results reveal that QKConv performs better than all unsupervised methods across three datasets and achieves competitive performance compared to supervised methods.