Knowledge Aware Conversation Generation with Explainable Reasoning over Augmented Graphs

By Zhibin Liu, Zheng-Yu Niu, Hua Wu, Haifeng Wang. 2019

103 citations recorded by the source corpus.

Two types of knowledge, triples from knowledge graphs and texts from documents, have been studied for knowledge aware open domain conversation generation, in which graph paths can narrow down vertex candidates for knowledge selection decision, and texts can provide rich information for response generation. Fusion of a knowledge graph and texts might yield mutually reinforcing advantages, but there is less study on that. To address this challenge, we propose a knowledge aware chatting machine with three components, an augmented knowledge graph with both triples and texts, knowledge selector, and knowledge aware response generator. For knowledge selection on the graph, we formulate it as a problem of multi-hop graph reasoning to effectively capture conversation flow, which is more explainable and flexible in comparison with previous works. To fully leverage long text information that differentiates our graph from others, we improve a state of the art reasoning algorithm with machine reading comprehension technology. We demonstrate the effectiveness of our system on two datasets in comparison with state-of-the-art models.

Subfields: Artificial Intelligence

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

Citations in the corpus

Knowledge Infused Learning (K-IL): Towards Deep Incorporation of Knowledge in Deep Learning (2019, background): “(Liu et al. 2019a) proposed K-BERT, that enriches the representations by injecting the triples from KGs into the sentence.”
A Survey on Knowledge Graphs: Representation, Acquisition, and Applications (2020, background): “[206] encoded knowledge to augment semantic representation and generated knowledge aware response by knowledge graph retrieval and graph attention mechanism under an encoderdecoder framework.”
Neural Path Hunter: Reducing Hallucination in Dialogue Systems via Path Grounding (2021, methodology): “KG’s have also risen in prominence in the context of dialogue models that propose to explicitly embed symbolic knowledge representations into a neural embedding space (Liu et al., 2019b; Zhu et al., 2017; Moon et al., 2019; Zhou et al., 2018; Xu et al., 2020).”
A Survey of Knowledge-enhanced Text Generation (2020, methodology): “e collection in here: https://github.com/wyu97/KENLG-Reading. Application Input Section 3 Section 4 Section 5 Topic Keyword LF KB KG DG KeyG GT Dialogue Sentence [158, 164] [102, 132] - [39, 93, 155] [108, 180] - - [36, 52, 169] system (i.e., utterance) [151, 179] [85, 144, 154] [138, 173] [96, 114,”
Identifying Untrustworthy Samples: Data Filtering for Open-domain Dialogues with Bayesian Optimization (2021, background): “, topics [25], sentence types [36], personas [19], emotions [43, 58], documents [30], or knowledge graphs [17, 27, 54, 59], augmenting the model itself [53, 57], or modifying the loss function [18].”