A Compare Aggregate Transformer for Understanding Document-grounded Dialogue

By Longxuan Ma, Weinan Zhang, Run-Quan Sun, Ting-Wei Liu. 2020

10 citations recorded by the source corpus.

Unstructured documents serving as external knowledge of the dialogues help to generate more informative responses. Previous research focused on knowledge selection (KS) in the document with dialogue. However, dialogue history that is not related to the current dialogue may introduce noise in the KS processing. In this paper, we propose a Compare Aggregate Transformer (CAT) to jointly denoise the dialogue context and aggregate the document information for response generation. We designed two different comparison mechanisms to reduce noise (before and during decoding). In addition, we propose two metrics for evaluating document utilization efficiency based on word overlap. Experimental results on the CMU_DoG dataset show that the proposed CAT model outperforms the state-of-the-art approach and strong baselines.

Subfields: Artificial Intelligence, Natural Language Processing, Model Architectures

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Natural Language Processing

In Model Architectures

Citations in the corpus

Retrieval Augmentation Reduces Hallucination in Conversation (2021, background): “…referred to as “knowledge selection” (Zhao et al., 2020b; Kim et al., 2020; Bruyn et al., 2020); learning how to attend to the relevant knowledge (Ma et al., 2020; Cai et al., 2020; Zhao et al., 2020a); or examining how much knowledge is present in large language models (Zhao et al., 2020c).”